U.S. patent number 11,263,749 [Application Number 17/339,151] was granted by the patent office on 2022-03-01 for predictive prognosis based on multimodal analysis.
This patent grant is currently assigned to In-Med Prognostics Inc.. The grantee listed for this patent is In-Med Prognostics Inc.. Invention is credited to Allen Richard Curran, Juhi Rajesh Desai, Hussain Murtuza Ghadiyali, Udit Goswami, Shivalika Goyal, Shubham Rajesh Halyal, Sonia Joy, Preeti Kabra, Praful Ramachandra Naik, Latha Chandrasekaran Poonamallee, Viyan Sathya Poonamallee, Rajesh Kumar Purushottam, Apeksha Sakegaonkar.
United States Patent |
11,263,749 |
Purushottam , et
al. |
March 1, 2022 |
Predictive prognosis based on multimodal analysis
Abstract
The present disclosure describes a method comprising: obtaining
one or more first images of a region of interest of an anatomy from
an image source; obtaining at least one of a text input, and one or
more physiological signals of a patient; automatically segmenting
one or more second images of at least one structure that resides
within the one or more first images; extracting one or more volumes
of the at least one structure from the one or more first images of
the region of interest; determining a feature associated with the
at least one structure based on the one or more volumes and one or
more inputs, and rendering the feature in at least one of a
two-dimensional (2D) format, a three-dimensional (3D) format, and
at least one anatomical plane.
Inventors: |
Purushottam; Rajesh Kumar
(Pune, IN), Curran; Allen Richard (Lewes, DE),
Poonamallee; Latha Chandrasekaran (Lewes, DE), Poonamallee;
Viyan Sathya (Lewes, DE), Desai; Juhi Rajesh (Pune,
IN), Naik; Praful Ramachandra (Pune, IN),
Kabra; Preeti (Pune, IN), Joy; Sonia (Lewes,
DE), Halyal; Shubham Rajesh (Pune, IN), Goswami;
Udit (Pune, IN), Sakegaonkar; Apeksha (Pune,
IN), Ghadiyali; Hussain Murtuza (Pune, IN),
Goyal; Shivalika (Pune, IN) |
Applicant: |
Name |
City |
State |
Country |
Type |
In-Med Prognostics Inc. |
Lewes |
DE |
US |
|
|
Assignee: |
In-Med Prognostics Inc. (Lewes,
DE)
|
Family
ID: |
80442662 |
Appl.
No.: |
17/339,151 |
Filed: |
June 4, 2021 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G16H
50/70 (20180101); G06T 7/12 (20170101); G16H
15/00 (20180101); G16H 30/20 (20180101); G16H
30/40 (20180101); G06N 3/08 (20130101); G16H
50/30 (20180101); G06F 16/55 (20190101); G16H
50/20 (20180101); G16H 10/60 (20180101); G06T
7/0012 (20130101); G16H 50/50 (20180101); G06T
2200/24 (20130101); G06T 2207/20084 (20130101); G06T
2207/30168 (20130101); G06T 2207/30016 (20130101); G06T
2207/20104 (20130101); G06T 2207/20081 (20130101) |
Current International
Class: |
G06K
9/00 (20060101); G16H 15/00 (20180101); G06N
3/08 (20060101); G06F 16/55 (20190101); G16H
50/50 (20180101); G06T 7/00 (20170101); G16H
30/20 (20180101); G16H 30/40 (20180101); G16H
10/60 (20180101); G16H 50/20 (20180101); G16H
50/70 (20180101); G16H 50/30 (20180101); G06T
7/12 (20170101) |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
101593345 |
|
Dec 2009 |
|
CN |
|
3714467 |
|
Sep 2020 |
|
EP |
|
Primary Examiner: Akhavannik; Hadi
Attorney, Agent or Firm: Dave Law Group LLC Dave; Raj S.
Claims
What is claimed is:
1. A method comprising: obtaining one or more first images of a
region of interest of an anatomy from an image source; obtaining at
least one of a text input, and one or more physiological signals of
a patient, wherein the text input comprises information of at least
one of micro-ethnicity information, an age, a race, a gender, a
medical condition, a symptom, clinical history, a patient history,
a medical test, medication information, and a cognitive analysis
report; automatically segmenting, through a neural network, one or
more second images of at least one structure that resides within
the one or more first images; extracting one or more volumes of the
at least one structure from the one or more first images of the
region of interest; determining a feature associated with the at
least one structure based on the one or more volumes and one or
more inputs; rendering the feature in at least one of a
two-dimensional (2D) format, a three-dimensional (3D) format, and
at least one anatomical plane, wherein the feature comprises at
least one of the one or more volumes of the region of interest
(ROI), a cortical thickness, an atrophy percentage, an asymmetry
index score, a subfield volumetry of the region of interest,
annular volume changes, a progressive supranuclear palsy (psp)
index score, a magnetic resonance perfusion imaging (MRPI) score, a
frontal horn width to intercaudate distance ratio (FH/CC), a medial
temporal lobe atrophy (MTA) score, a global cortical atrophy (GCA)
scale, identification of Intracranial bleeds, hemorrhage,
microbleeds and their volume analysis, a fracture detection, a
midline shift identification, a measurement of the midline shift
identification and the at least one structure with respect to the
midline shift identification, identifying a pathology associated
with the at least one structure, classifying the pathology
identified, a tissue density identification, an infarct
identification, a Penumbra-core-viable tissue identification,
classification and volume calculation, diffusion-weighted imaging
(DWI) maps and apparent diffusion coefficient (ADC) maps of the at
least one structure, perfusion maps comprising resting state
functional magnetic resonance imaging (rsfMRI), an alberta stroke
programme early CT score (ASPECTS) calculation, a collateral
detection, a mismatch ratio calculation, an angiography labeling
and/or annotation, a large vessel occlusion (LVO) detection, an
Hypoperfusion index calculation, Diffusion tensor imaging (DTI)
fiber tracks, neural pathway connectivity maps, correlation between
a signal input, an image input and the text input, classifying the
signal input, identifying a normal signal, identifying an abnormal
signal, identifying a pre-ictal signal, identifying an ictal
signal, extracting symptoms, and grading of condition specific
effects; transforming automatically the one or more physiological
signals from a sinusoidal wave format to a quantitative format,
wherein the quantitative format comprises a numerical
representation of the one or more physiological signals; predicting
a prognosis based on correlation with the image input and
integrated analysis of at least one spike detected, and the
numerical representation of the one or more physiological signals;
and generating an analysis report based on the prognosis, wherein
the analysis report comprising a snippet describing the prognosis,
the one or more volumes of the at least one structure, one or more
quantitative volumes, and a graphical representation of the
prognosis.
2. The method of claim 1, wherein automatically segmenting, through
the neural network, the one or more second images of the at least
one structure that resides within the one or more first images
comprise: performing a second quality analysis manually on the one
or more second images that are segmented; and determining whether
the one or more second images, that are segmented, passes the
second quality analysis.
3. The method of claim 2, wherein determining whether the one or
more second images, that are segmented, passes the second quality
analysis comprises: providing a user interface when the one or more
second images that are segmented fails the second quality analysis;
manually editing and correcting at least one of boundaries and the
one or more volumes of the at least one structure based on one or
more inputs received; and creating a mask for the at least one
structure.
4. The method of claim 1, wherein automatically segmenting, through
the neural network, the one or more second images of the at least
one structure that resides within the one or more first images
comprise: training the neural network using at least one of (a) the
one or more first images, (b) the information of at least one of
the micro-ethnicity information, the age, the race, the gender, the
medical condition, the symptom, the clinical history, the patient
history, the medical test, the medication information, and the
cognitive analysis report, (c) the one or more physiological
signals, (d) the one or more volumes of the at least one structure,
(e) one or more reference volumes, and (f) one or more reference
segmented second images.
5. The method of claim 3, wherein manually editing and correcting
at least one of the boundaries and the one or more volumes of the
at least one structure based on the one or more inputs received
comprises: creating a log for the mask using the one or more inputs
received; retraining the neural network based on the log created;
and automatically segmenting, through the neural network, the one
or more second images of the at least one structure in future based
on the retraining provided to the neural network.
6. The method of claim 1, wherein extracting the one or more
volumes of the at least one structure from the one or more first
images of the region of interest comprises: assigning a voxel of a
mask of the one or more second images, that are segmented, as a
unit; tabulating a plurality of units in the mask; and estimating
one or more quantitative volumes of the at least one structure from
the plurality of units.
7. The method of claim 1, further comprising: recording the one or
more volumes of the at least one structure in a database; and
categorizing the one or more volumes of the at least one structure
in the database with respect to one or more categories of at least
one of the micro-ethnicity information, an intracranial volume
(ICV), the age, the race, the gender, a family history, the
clinical history, the patient history, the symptom, psych analysis
information, brain dominance information, food habitat information,
stress information, and the medical condition.
8. The method of claim 1, wherein extracting the one or more
volumes of the at least one structure from the one or more first
images of the region of interest: extracting one or more boundaries
of the at least one structure from the one or more first images;
and populating one or more voxels within the one or more boundaries
of the at least one structure using one or more identifiers.
9. The method of claim 3, wherein manually editing and correcting
at least one of the boundaries, and the one or more volumes of the
at least one structure based on the one or more inputs received
comprises: performing at least one of adding, and deleting one or
more voxels within the boundaries of the at least one structure
based on the one or more inputs received.
10. The method of claim 1, further comprising: detecting at least
one spike within the one or more physiological signals that
indicates abnormality; and predicting the prognosis based on
correlation and integrated analysis of the at least one spike
detected, the text input, and the one or more volumes.
11. The method of claim 10, further comprising: correlating with at
least one of temporal resolution and spatial resolution of the
image input and detecting an abnormal region, using the neural
network, in the one or more volumes based on the at least one spike
detected; and indicating the abnormal region using a different
identifier.
12. The method of claim 10, further comprising: detecting an
abnormal region, using the neural network, in the one or more
physiological signals based on volumetric analysis; and indicating
the abnormal region, comprising the at least one spike, using a
different identifier.
13. The method of claim 4, wherein the one or more reference
volumes range between 25th and 95th percentile, wherein the 25th
and the 95th percentile are calculated by matching at least one of
the age, the gender, the micro-ethnicity information, and an
intracranial volume (ICV) of the patient with a normative
population of individuals and then deriving the 25th and the 95th
percentile references.
14. The method of claim 13, wherein the 25th and the 95th
percentile is calculated by matching the medical condition of the
patient with a population of individuals having the medical
condition and then deriving the 25th and the 95th percentile.
15. The method of claim 14, further comprising: predicting a first
prognosis state of the patient based at least one of the medical
condition, and first medication information of the patient at a
first point of time and generating a first analysis report; and
predicting a second prognosis state of the patient based on at
least one of the medical condition, and second medication
information of the patient at a second point of time and generating
a second analysis report.
16. The method of claim 15, further comprising: comparing the first
prognosis state and the second prognosis state; determining a
percentage of one of a deterioration and an improvement in at least
one of the one or more volumes, and one or more quantitative
volumes based on comparison of the first prognosis state and the
second prognosis state; and training, the neural network, using at
least one of medical condition, the first medication information,
the second medication information, and the percentage of the
deterioration or the improvement in at least one of the one or more
volumes, and the one or more quantitative volumes at a plurality of
different points of time.
17. The method of claim 16, further comprising: detecting a
diagnosis, via the neural network, at a third point of time by
comparing the first prognosis state and the second prognosis state
based on the training; performing a predictive prognosis, via the
neural network, and predicting a third prognosis state of the
patient at the third point of time based on the training; and
generating a third analysis report comprising a clinical analytical
outcome at the third point of time.
18. A system comprising: a server comprising a memory, and a
processor communicatively coupled to the memory, the processor
operable to obtain one or more first images of a region of interest
of an anatomy from an image source; obtain at least one of a text
input, and one or more physiological signals of a patient, wherein
the text input comprises information of at least one of
micro-ethnicity information, an age, a race, a gender, a medical
condition, a symptom, clinical history, a patient history, a
medical test, medication information, and a cognitive analysis
report; automatically segment, through a neural network, one or
more second images of at least one structure that resides within
the one or more first images; extract one or more volumes of the at
least one structure from the one or more first images of the region
of interest; determine a feature associated with the at least one
structure based on the one or more volumes and one or more inputs;
render the feature in at least one of a two-dimensional (2D)
format, a three-dimensional (3D) format, and at least one
anatomical plane, wherein the feature comprises at least one of the
one or more volumes of the region of interests (ROI), a cortical
thickness, an atrophy percentage, an asymmetry index score, a
subfield volumetry of the region of interest, annular volume
changes, a progressive supranuclear palsy (psp) index score, a
magnetic resonance perfusion imaging (MRPI) score, a frontal horn
width to intercaudate distance ratio (FH/CC), a medial temporal
lobe atrophy (MTA) score, a global cortical atrophy (GCA) scale,
identification of Intracranial bleeds, hemorrhage, microbleeds and
their volume analysis, a fracture detection, a midline shift
identification, a measurement of the midline shift identification
and the at least one structure with respect to the midline shift
identification, identifying a pathology associated with the at
least one structure, classifying the pathology identified, a tissue
density identification, an infarct identification, a
Penumbra-core-viable tissue identification, classification and
volume calculation, diffusion-weighted imaging (DWI) maps and
apparent diffusion coefficient (ADC) maps of the at least one
structure, perfusion maps comprising resting state functional
magnetic resonance imaging (rsfMRI), an alberta stroke programme
early CT score (ASPECTS) calculation, a collateral detection, a
mismatch ratio calculation, an angiography labeling and/or
annotation, a large vessel occlusion (LVO) detection, an
Hypoperfusion index calculation, Diffusion tensor imaging (DTI)
fiber tracks, neural pathway connectivity maps, correlation between
a signal input, an image input and the text input, classifying the
signal input, identifying a normal signal, identifying an abnormal
signal, identifying a pre-ictal signal, identifying an ictal
signal, extracting symptoms, and grading of condition specific
effects; transform automatically the one or more physiological
signals from a sinusoidal wave format to a quantitative format,
wherein the quantitative format comprises a numerical
representation of the one or more physiological signals; predict a
prognosis based on correlation with the image input and integrated
analysis of at least one spike detected, and the numerical
representation of the one or more physiological signals; and
generate an analysis report based on the prognosis, wherein the
analysis report comprising a snippet describing the prognosis, the
one or more volumes of the at least one structure, one or more
quantitative volumes, and a graphical representation of the
prognosis.
19. The system of claim 18, wherein the processor operable to
detect at least one spike within the one or more physiological
signals that indicates abnormality; and predict the prognosis based
on correlation and integrated analysis of the at least one spike
detected, the text input, and the one or more volumes.
20. A non-transitory computer storage medium storing a sequence of
instructions, which when executed by a processor, causes: obtaining
one or more first images of a region of interest of an anatomy from
an image source; obtaining at least one of a text input, and one or
more physiological signals of a patient, wherein the text input
comprises information of at least one of micro-ethnicity
information, an age, a race, a gender, a medical condition, a
symptom, clinical history, a patient history, a medical test,
medication information, and a cognitive analysis report;
automatically segmenting, through a neural network, one or more
second images of at least one structure that resides within the one
or more first images; extracting one or more volumes of the at
least one structure from the one or more first images of the region
of interest; determining a feature associated with the at least one
structure based on the one or more volumes and one or more inputs;
rendering the feature in at least one of a two-dimensional (2D)
format, a three-dimensional (3D) format, and at least one
anatomical plane, wherein the feature comprises at least one of the
one or more volumes of the region of interest (ROI), a cortical
thickness, an atrophy percentage, an asymmetry index score, a
subfield volumetry of the region of interest, annular volume
changes, a progressive supranuclear palsy (psp) index score, a
magnetic resonance perfusion imaging (MRPI) score, a frontal horn
width to intercaudate distance ratio (FH/CC), a medial temporal
lobe atrophy (MTA) score, a global cortical atrophy (GCA) scale,
identification of Intracranial bleeds, hemorrhage, microbleeds and
their volume analysis, a fracture detection, a midline shift
identification, a measurement of the midline shift identification
and the at least one structure with respect to the midline shift
identification, identifying a pathology associated with the at
least one structure, classifying the pathology identified, a tissue
density identification, an infarct identification, a
Penumbra-core-viable tissue identification, classification and
volume calculation, diffusion-weighted imaging (DWI) maps and
apparent diffusion coefficient (ADC) maps of the at least one
structure, perfusion maps comprising resting state functional
magnetic resonance imaging (rsfMRI), an alberta stroke programme
early CT score (ASPECTS) calculation, a collateral detection, a
mismatch ratio calculation, an angiography labeling and/or
annotation, a large vessel occlusion (LVO) detection, an
Hypoperfusion index calculation, Diffusion tensor imaging (DTI)
fiber tracks, neural pathway connectivity maps, correlation between
a signal input, an image input and the text input, classifying the
signal input, identifying a normal signal, identifying an abnormal
signal, identifying a pre-ictal signal, identifying an ictal
signal, extracting symptoms, and grading of condition specific
effects; transforming automatically the one or more physiological
signals from a sinusoidal wave format to a quantitative format,
wherein the quantitative format comprises a numerical
representation of the one or more physiological signals; predicting
a prognosis based on correlation with the image input and
integrated analysis of at least one spike detected, and the
numerical representation of the one or more physiological signals;
and generating an analysis report based on the prognosis, wherein
the analysis report comprising a snippet describing the prognosis,
the one or more volumes of the at least one structure, one or more
quantitative volumes, and a graphical representation of the
prognosis.
Description
FIELD OF THE INVENTION
The present disclosure relates broadly to computer aided prognosis,
and more particularly to predictive prognosis based on multimodal
and multivariate pattern analysis of image, text, and signal
inputs.
BACKGROUND
"In the past, 2D models were the main models for medical image
processing applications, whereas the wide adoption of 3D models has
appeared only in recent years. The 2D Fuzzy C-Means (FCM) algorithm
has been extensively used for segmenting medical images due to its
effectiveness. Various extensions of it were proposed throughout
the years. In this work, we propose a modified version of FCM for
segmenting 3D medical volumes, which has been rarely implemented
for 3D medical image segmentation. We present a parallel
implementation of the proposed algorithm using Graphics Processing
Unit (GPU). Researchers state that efficiency is one of the main
problems of using FCM for medical imaging when dealing with 3D
models. Thus, a hybrid parallel implementation of FCM for
extracting volume objects from medical files is proposed. The
proposed algorithm has been validated using real medical data and
simulated phantom data. Segmentation accuracy of predefined
datasets and real patient datasets were the key factors for the
system validation. The processing times of both the sequential and
the parallel implementations are measured to illustrate the
efficiency of each implementation. The acquired results conclude
that the parallel implementation is 5.times. faster than the
sequential version of the same operation." [Source: Parallel
Implementation for 3D Medical Volume Fuzzy Segmentation; Shadi
AlZfdesu'bi; Mohammed A. Shehab; Mahmoud Al-Ayyoub; Yaser Jararweh;
Brij Gupta; published in July 2018].
"Neuroimaging has been playing pivotal roles in clinical diagnosis
and basic biomedical research in the past decades. As described in
the following section, the most widely used imaging modalities are
magnetic resonance imaging (MRI), computerized tomography (CT),
positron emission tomography (PET), and single-photon emission
computed tomography (SPECT). Among them, MRI itself is a
non-radioactive, non-invasive, and versatile technique that has
derived many unique imaging modalities, such as diffusion-weighted
imaging, diffusion tensor imaging, susceptibility-weighted imaging,
and spectroscopic imaging. PET is also versatile, as it may use
different radiotracers to target different molecules or to trace
different biologic pathways of the receptors in the body." [Source:
Advances in multimodal data fusion in neuroimaging: Overview,
challenges, and novel orientation; Yu-Dong Zhang; Zhengchao Dong;
Shui-Hua Wang; Xiang Yu; Xujing Yao; Qinghua Zhou; Hua Hu, Min Li;
Carmen Jimenez-Mesa; Javier Ramirez; Francisco J. Martinez; and
Juan Manuel Gorriz; published on Jul. 17, 2020]
"Therefore, these individual imaging modalities (the use of one
imaging modality), with their characteristics in signal sources,
energy levels, spatial resolutions, and temporal resolutions,
provide complementary information on anatomical structure,
pathophysiology, metabolism, structural connectivity, functional
connectivity, etc. Over the past decades, everlasting efforts have
been made in developing individual modalities and improving their
technical performance. Directions of improvements include data
acquisition and data processing aspects to increase spatial and/or
temporal resolutions, improve signal-to-noise ratio and contrast to
noise ratio, and reduce scan time. On application aspects,
individual modalities have been widely used to meet clinical and
scientific challenges. At the same time, technical developments and
biomedical applications of the concert, integrated use of multiple
neuroimaging modalities is trending up in both research and
clinical institutions. The driving force of this trend is twofold.
First, all individual modalities have their limitations. For
example, some lesions in MS can appear normal in T1-weighted or
T2-weighted MR images but show pathological changes in DWI or SWI
images.sup.[1]. Second, a disease, disorder, or lesion may manifest
itself in different forms, symptoms, or etiology; or on the other
hand, different diseases may share some common symptoms or
appearances.sup.[2, 3]. Therefore, an individual image modality may
not be able to reveal a complete picture of the disease; and
multimodal imaging modality (the use of multiple imaging
modalities) may lead to a more comprehensive understanding,
identify factors, and develop biomarkers of the disease." [Source:
Advances in multimodal data fusion in neuroimaging: Overview,
challenges, and novel orientation; Yu-Dong Zhang; Zhengchao Dong;
Shui-Hua Wang; Xiang Yu; Xujing Yao; Qinghua Zhou; Hua Hu, Min Li;
Carmen Jimenez-Mesa; Javier Ramirez; Francisco J. Martinez; and
Juan Manuel Gorriz; published on Jul. 17, 2020].
Considering the knowledge of the persons skilled in the art, there
is a long-felt need for a structural analysis and integrated
multimodal analysis of image, signal, and text inputs for ensuring
accuracy in atrophy determination, clinical prognosis, and
diagnosis.
SUMMARY
The present disclosure describes one or more aspects of image
segmentation, volumetric extraction and volumetric analysis for
performing at least one of predictive prognosis, diagnosis, and
atrophy determination.
In an aspect, a method is described herein. The method comprises:
obtaining one or more first images of a region of interest of an
anatomy from an image source; obtaining at least one of a text
input, and one or more physiological signals of a patient;
automatically segmenting, through a neural network, one or more
second images of at least one structure that resides within the one
or more first images; extracting one or more volumes of the at
least one structure from the one or more first images of the region
of interest; determining a feature associated with the at least one
structure based on the one or more volumes and one or more inputs;
and rendering the feature in at least one of a two-dimensional (2D)
format, a three-dimensional (3D) format, and at least one
anatomical plane. The text input comprises information of at least
one of micro-ethnicity information, an age, a race, a gender, a
medical condition, a symptom, clinical history, a patient history,
a medical test, medication information, and a cognitive analysis
report. The feature comprises at least one of the one or more
volumes of the region of interest (ROI), a cortical thickness, an
atrophy percentage, an asymmetry index score, a subfield volumetry
of the region of interest, annular volume changes, a progressive
supranuclear palsy (psp) index score, a magnetic resonance
perfusion imaging (MRPI) score, a frontal horn width to
intercaudate distance ratio (FH/CC), a medial temporal lobe atrophy
(MTA) score, a global cortical atrophy (GCA) scale, identification
of Intracranial bleeds, hemorrhage, microbleeds and their volume
analysis, a fracture detection, a midline shift identification, a
measurement of the midline shift identification and the at least
one structure with respect to the midline shift identification,
identifying a pathology associated with the at least one structure,
classifying the pathology identified, a tissue density
identification, an infarct identification, a Penumbra-core-viable
tissue identification, classification and volume calculation,
diffusion-weighted imaging (DWI) maps and apparent diffusion
coefficient (ADC) maps of the at least one structure, perfusion
maps comprising resting state functional magnetic resonance imaging
(rsfMRI), an alberta stroke programme early CT score (ASPECTS)
calculation, a collateral detection, a mismatch ratio calculation,
an angiography labeling and/or annotation, a large vessel occlusion
(LVO) detection, an Hypoperfusion index calculation, Diffusion
tensor imaging (DTI) fiber tracks, neural pathway connectivity
maps, correlation between a signal input, an image input and the
text input, classifying the signal input, identifying a normal
signal, identifying an abnormal signal, identifying a pre-ictal
signal, identifying an ictal signal, extracting symptoms, and
grading of condition specific effects.
In an embodiment, the method further comprises performing a first
quality analysis on the one or more first images of the region of
interest prior to automatically segmenting the at least one
structure.
In another embodiment, performing the first quality analysis on the
one or more first images of the region of interest prior to
automatically segmenting the one or more second images of the at
least one structure comprises: determining whether the one or more
first images of the region of interest are obtained from one of
computed tomography (CT), positron emission tomography (PET),
structural magnetic resonance imaging (sMRT), functional magnetic
resonance imaging (fMRI), Diffusion-weighted imaging (DWI),
Diffusion Tensor Imaging (DTI), and magnetic resonance imaging
(MRI) with a predefined magnetic strength value.
In yet another embodiment, the predefined magnetic strength value
comprises a value more than 1.5 Tesla.
In yet another embodiment, automatically segmenting, through the
neural network, the one or more second images of the at least one
structure that resides within the one or more first images
comprise: performing a second quality analysis manually on the one
or more second images that are segmented; and determining whether
the one or more second images, that are segmented, passes the
second quality analysis.
In yet another embodiment, determining whether the one or more
second images, that are segmented, passes the second quality
analysis comprises: providing a user interface when the one or more
second images, that are segmented, fails the second quality
analysis; manually editing and correcting at least one of
boundaries and the one or more volumes of the at least one
structure based on one or more inputs received from the user; and
creating a mask for the at least one structure.
In yet another embodiment, the image source comprises one of (a) a
magnetic resonance imaging (MRI) machine, (b) a computed tomography
(CT) machine, and (c) a computing unit.
In yet another embodiment, the anatomy belongs to an organism.
In yet another embodiment, the organism comprises one of (a) a
human being, (b) an animal, (c) a mammal, and (d) a bird.
In yet another embodiment, the computing unit comprises a personal
digital assistant.
In yet another embodiment, automatically segmenting, through the
neural network, the one or more second images of the at least one
structure that resides within the one or more first images
comprises: training the neural network using at least one of (a)
the one or more first images, (b) the information of at least one
of the micro-ethnicity information, the age, the race, the gender,
the medical condition, the symptom, the clinical history, the
patient history, the medical test, the medication information, and
the cognitive analysis report, (c) the one or more physiological
signals, (d) the one or more volumes of the at least one structure,
(e) one or more reference volumes, and (f) one or more reference
segmented second images.
In yet another embodiment, manually editing and correcting at least
one of the boundaries and the one or more volumes of the at least
one structure based on the one or more inputs received from the
user comprises: creating a log for the mask using the one or more
inputs received from the user; retraining the neural network based
on the log created; and automatically segmenting, through the
neural network, the one or more second images of the at least one
structure in future based on the retraining provided to the neural
network.
In yet another embodiment, extracting the one or more volumes of
the at least one structure from the one or more first images of the
region of interest comprises: assigning a voxel of a mask of the
one or more second images, that are segmented, as a unit;
tabulating a plurality of units in the mask; and estimating one or
more quantitative volumes of the at least one structure from the
plurality of units.
In yet another embodiment, the method further comprises: recording
the one or more volumes of the at least one structure in a
database; and categorizing the one or more volumes of the at least
one structure in the database with respect to one or more
categories of at least one of the micro-ethnicity information, an
intracranial volume (ICV), the age, the race, the gender, a family
history, the clinical history, the patient history, the symptom,
psych analysis information, brain dominance information, food
habitat information, stress information, and the medical
condition.
In yet another embodiment, extracting the one or more volumes of
the at least one structure from the one or more first images of the
region of interest: extracting one or more boundaries of the at
least one structure from the one or more first images; and
populating one or more voxels within the one or more boundaries of
the at least one structure using one or more identifiers.
In yet another embodiment, obtaining the one or more first images
of the region of interest of the anatomy from the image source
comprises: obtaining the one or more first images of the region of
interest in a Digital Imaging and Communications in Medicine
(DICOM) format.
In yet another embodiment, the method further comprises anonymizing
the one or more first images by discarding metadata from the one or
more first images. The metadata comprises user identifying
information.
In yet another embodiment, the method further comprises discarding
the metadata from the one or more first images by converting the
one or more first images from a Digital Imaging and Communications
in Medicine (DICOM) format to a Neuroimaging Informatics Technology
Initiative (NIfTI) format.
In yet another embodiment, the one or more physiological signals
comprises at least one of an event related potential (ERP),
electrocardiography (ECG) signal, an electroencephalogram (EEG)
signal, an Electromyography (EMG), a galvanic skin response (GSR),
a blood pressure, and a pulse rate.
In yet another embodiment, the one or more first images comprise
one of a three-dimensional (3D) magnetic resonance imaging (MRI), a
three-dimensional (3D) computed tomography (CT), a
three-dimensional (3D) Functional magnetic resonance imaging
(fMRI), and a three-dimensional (3D) positron emission tomography
(PET).
In yet another embodiment, the method further comprises: assigning
a user identification data to the patient upon obtaining at least
one of the one or more first images, the text input, and the one or
more physiological signals. The text input comprises at least one
of the patient history, and a cognitive test.
In yet another embodiment, assigning the user identification data
to the patient upon obtaining at least one of the one or more first
images, the text input, and the one or more physiological signals
comprises: assigning a first user identification data to the
patient upon obtaining at least one of the one or more first
images, the text input, and the one or more physiological signals
at a first station; and assigning a second user identification data
to the patient upon obtaining at least one of the one or more first
images, the text input, and the one or more physiological signals
at a second station.
In yet another embodiment, the method further comprises linking
information of the patient associated with the first user
identification data and the second user identification data upon
receiving a linking request from the user.
In yet another embodiment, manually editing and correcting at least
one of the boundaries, and the one or more volumes of the at least
one structure based on the one or more inputs received from the
user comprises: performing at least one of adding, and deleting one
or more voxels within the boundaries of the at least one structure
based on the one or more inputs received from the user.
In yet another embodiment, the at least one structure comprises at
least one organ.
In yet another embodiment, the at least one organ comprises a body
part of at least one of one of a circulatory system, a nervous
system, a muscular system, an endocrine system, a respiratory
system, a digestive system, a urinary system, a reproductive
system, an integumentary system, an immune system, and a skeletal
system.
In yet another embodiment, the at least one anatomical plane
comprises a sagittal plane, an axial plane, a parasagittal plane,
and a coronal plane.
In another aspect, a method is described herein. The method
comprises: obtaining one or more first images of a region of
interest of an anatomy from an image source; obtaining at least one
of a text input, and one or more physiological signals of a
patient; automatically segmenting one or more second images of at
least one structure that resides within the one or more first
images; estimating one or more quantitative volumes of the at least
one structure; and predicting a prognosis based on comparison of
the one or more quantitative volumes of the at least one structure
with one or more reference volumes, at least one of the text input,
and the one or more physiological signals. The text input comprises
information of at least one of micro-ethnicity information, an age,
a race, a gender, a medical condition, a symptom, clinical history,
a patient history, a medical test, medication information, and a
cognitive analysis report.
In an embodiment, the method further comprises generating a
structure-based analysis report comprising at least one of the one
or more quantitative volumes of the at least one structure, a
snippet of output, a graphical representation of the prognosis, and
the one or more second images in at least one anatomical plane.
In another embodiment, the one or more reference volumes range
between 25th and 95th percentile. The 25th and the 95th percentile
are calculated by matching at least one of an age, a gender, a
micro-ethnicity information, and an intracranial volume (ICV) of
the patient with a normative population of individuals and then
deriving the 25th and the 95th percentile references.
In yet another embodiment, the 25th and the 95th percentile is
calculated by matching a medical condition of the patient with a
population of individuals having the medical condition and then
deriving the 25th and the 95th percentile.
In yet another embodiment, the method comprises: predicting a first
prognosis state of the patient based at least one of the medical
condition, and a first medication information of the patient at a
first point of time and generating a first analysis report; and
predicting a second prognosis state of the patient based on at
least one of the medical condition, and a second medication
information of the patient at a second point of time and generating
a second analysis report.
In yet another embodiment, the method comprises: comparing the
first prognosis state and the second prognosis state; and
determining a percentage of one of a deterioration and an
improvement in at least one of one or more volumes, and the one or
more quantitative volumes based on the comparison of the first
prognosis state and the second prognosis state.
In yet another embodiment, the method comprises: training, a neural
network, using at least one of medical condition, the first
medication information, the second medication information, and the
percentage of the deterioration or the improvement in at least one
of one or more volumes, and the one or more quantitative volumes at
a plurality of different point of times.
In yet another embodiment, the method further comprises: detecting
a diagnosis of the patient at a third point of time; performing a
predictive prognosis and predicting a third prognosis state of the
patient at the third point of time; and generating a third analysis
report comprising a clinical analytical outcome at the third point
of time.
In yet another embodiment, the method further comprises: rendering
the third analysis report to a physician, the third analysis report
comprises brief summary that assist the physician in determining
whether a first medical regime prescribed to a patient is
effective, and prescribing a second medication regime with respect
to the third prognosis state of the patient.
In yet another embodiment, obtaining the one or more first images
of the region of interest of the anatomy from the image source
comprises: obtaining the one or more first images of the region of
interest at a first instance; and obtaining the one or more first
images of the region of interest at a second instance.
In yet another embodiment, the method further comprises generating
a first structure-based analysis report based on the one or more
first images obtained at the first instance; and generating a
second structure-based analysis report based on the one or more
first images obtained at the second instance.
In yet another embodiment, the method further comprises predicting
the prognosis based on comparison of the first structure-based
analysis report and the second structure-based analysis report, and
the one or more first images of the region of interest that are
obtained at a third instance; estimating one of a progression and a
regression of the prognosis associated with the at least one
structure between the first instance and the second instance; and
generating a third structure-based analysis report comprising at
least one of the one or more quantitative volumes of the at least
one structure, a snippet, a graphical representation of the
prognosis, and the one or more second images in at least one
anatomical plane.
In yet another aspect, a method is described herein. The method
comprises: obtaining one or more first images of a region of
interest of an anatomy from an image source; obtaining a text
input; automatically segmenting, through a neural network, one or
more second images of at least one structure that resides within
the one or more first images; extracting one or more volumes of the
at least one structure from the one or more first images of the
region of interest; determining a feature associated with the at
least one structure based on the one or more volumes and one or
more inputs; and rendering the feature in at least one of a
two-dimensional (2D) format, a three-dimensional (3D) format, and
at least one anatomical plane. The text input comprises
micro-ethnicity information of a patient. The feature comprises at
least one of the one or more volumes of the region of interest
(ROI), a cortical thickness, an atrophy percentage, an asymmetry
index score, a subfield volumetry of the region of interest,
annular volume changes, a progressive supranuclear palsy (psp)
index score, a magnetic resonance perfusion imaging (MRPI) score, a
frontal horn width to intercaudate distance ratio (FH/CC), a medial
temporal lobe atrophy (MTA) score, a global cortical atrophy (GCA)
scale, identification of Intracranial bleeds, hemorrhage,
microbleeds and their volume analysis, a fracture detection, a
midline shift identification, a measurement of the midline shift
identification and the at least one structure with respect to the
midline shift identification, identifying a pathology associated
with the at least one structure, classifying the pathology
identified, a tissue density identification, an infarct
identification, a Penumbra-core-viable tissue identification,
classification and volume calculation, diffusion-weighted imaging
(DWI) maps and apparent diffusion coefficient (ADC) maps of the at
least one structure, perfusion maps comprising resting state
functional magnetic resonance imaging (rsfMRI), an alberta stroke
programme early CT score (ASPECTS) calculation, a collateral
detection, a mismatch ratio calculation, an angiography labeling
and/or annotation, a large vessel occlusion (LVO) detection, an
Hypoperfusion index calculation, Diffusion tensor imaging (DTI)
fiber tracks, neural pathway connectivity maps, correlation between
a signal input, an image input and the text input, classifying the
signal input, identifying a normal signal, identifying an abnormal
signal, identifying a pre-ictal signal, identifying an ictal
signal, extracting symptoms, and grading of condition specific
effects.
In yet another embodiment, automatically segmenting, through the
neural network, the one or more second images of the at least one
structure that resides within the one or more first images
comprises: training the neural network using at least one of (a)
the one or more first images, (b) the information of at least one
of the micro-ethnicity information, a cognitive score, and a
patient history, (c) the one or more volumes, (d) one or more
reference volumes, and (e) one or more reference segmented second
images.
In yet another embodiment, obtaining the text input of the patient
comprises: obtaining the micro-ethnicity information of the patient
through a global positioning system (GPS).
In yet another embodiment, obtaining the text input of the patient
comprises: extracting at least one of the text inputs of the
patient, a cognitive score, and detailed history of the patient
from one or more patient records available in one or more
databases.
In yet another embodiment, the text input further comprises an age,
a race, and a gender.
In yet another aspect, a database is described herein. The database
comprises one or more first images of a region of interest of an
anatomy obtained from an image source; a text input comprising
information of at least one of micro-ethnicity information, an age,
a race, a gender, a medical condition, a symptom, clinical history,
a patient history, a medical test, medication information, and a
cognitive analysis report; one or more physiological signals
acquired from a patient; one or more volumes of at least one
structure that resides within the one or more first images with
respect to micro-ethnicity information in at least one of a
three-dimensional (3d) format, and at least one anatomical plane;
one or more quantitative volumes of the at least one structure of
the region of interest that are categorized with respect to the
micro-ethnicity information; one or more structure-based analysis
report generated based on at least one of the one or more first
images, the text input, the one or more quantitative volumes of the
at least one structure, and the one or more physiological signals;
one or more reference volumes; and an index for the one or more
volumes, and the one or more quantitative volumes. The one or more
volumes, the one or more quantitative volumes, and the one or more
reference volumes are stored in a data structure on a computer
readable storage medium that is associated with a computer
executable program code.
In an embodiment the database comprises: user identification data
assigned to the patient; a progression and a regression state of
prognosis; and a health condition of the patient.
In yet another aspect, a system is described herein. The system
comprises a server comprising a memory, and a processor
communicatively coupled to the memory. The processor is operable to
obtain one or more first images of a region of interest of an
anatomy from an image source; obtain at least one of a text input,
and one or more physiological signals of a patient; automatically
segment, through a neural network, one or more second images of at
least one structure that resides within the one or more first
images; extract one or more volumes of the at least one structure
from the one or more first images of the region of interest;
determine a feature associated with the at least one structure
based on the one or more volumes and one or more inputs; and render
the feature in at least one of a two-dimensional (2D) format, a
three-dimensional (3D) format, and at least one anatomical plane.
The text input comprises information of at least one of
micro-ethnicity information, an age, a race, a gender, a medical
condition, a symptom, clinical history, a patient history, a
medical test, medication information, and a cognitive analysis
report. The feature comprises at least one of the one or more
volumes of the region of interest (ROI), a cortical thickness, an
atrophy percentage, an asymmetry index score, a subfield volumetry
of the region of interest, annular volume changes, a progressive
supranuclear palsy (psp) index score, a magnetic resonance
perfusion imaging (MRPI) score, a frontal horn width to
intercaudate distance ratio (FH/CC), a medial temporal lobe atrophy
(MTA) score, a global cortical atrophy (GCA) scale, identification
of Intracranial bleeds, hemorrhage, microbleeds and their volume
analysis, a fracture detection, a midline shift identification, a
measurement of the midline shift identification and the at least
one structure with respect to the midline shift identification,
identifying a pathology associated with the at least one structure,
classifying the pathology identified, a tissue density
identification, an infarct identification, a Penumbra-core-viable
tissue identification, classification and volume calculation,
diffusion-weighted imaging (DWI) maps and apparent diffusion
coefficient (ADC) maps of the at least one structure, perfusion
maps comprising resting state functional magnetic resonance imaging
(rsfMRI), an alberta stroke programme early CT score (ASPECTS)
calculation, a collateral detection, a mismatch ratio calculation,
an angiography labeling and/or annotation, a large vessel occlusion
(LVO) detection, an Hypoperfusion index calculation, Diffusion
tensor imaging (DTI) fiber tracks, neural pathway connectivity
maps, correlation between a signal input, an image input and the
text input, classifying the signal input, identifying a normal
signal, identifying an abnormal signal, identifying a pre-ictal
signal, identifying an ictal signal, extracting symptoms, and
grading of condition specific effects.
In yet another aspect, a system is described herein. The system
comprises a server comprising a memory, and a processor
communicatively coupled to the memory. The processor is operable to
obtain one or more first images of a region of interest of an
anatomy from an image source; obtain at least one of a text input,
and one or more physiological signals of a patient; automatically
segment one or more second images of at least one structure that
resides within the one or more first images; estimate one or more
quantitative volumes of the at least one structure; and predict a
prognosis based on comparison of the one or more quantitative
volumes of the at least one structure with one or more reference
volumes, the text input, and the one or more physiological signals.
The text input comprises information of at least one of
micro-ethnicity information, an age, a race, a gender, a medical
condition, a symptom, clinical history, a patient history, a
medical test, medication information, and a cognitive analysis
report.
In yet another aspect, a system is described herein. The system
comprises a server comprising a memory, and a processor
communicatively coupled to the memory. The processor is operable to
obtain one or more first images of a region of interest of an
anatomy from an image source; obtain a text input; automatically
segment, through a neural network, one or more second images of at
least one structure that resides within the one or more first
images; extract one or more volumes of the at least one structure
from the one or more first images of the region of interest;
determine a feature associated with the at least one structure
based on the one or more volumes and one or more inputs; and render
the feature in at least one of a two-dimensional (2D) format, a
three-dimensional (3D) format, and at least one anatomical plane.
The text input comprises micro-ethnicity information of a patient.
The feature comprises at least one of the one or more volumes of
the region of interest (ROI), a cortical thickness, an atrophy
percentage, an asymmetry index score, a subfield volumetry of the
region of interest, annular volume changes, a progressive
supranuclear palsy (psp) index score, a magnetic resonance
perfusion imaging (MRPI) score, a frontal horn width to
intercaudate distance ratio (FH/CC), a medial temporal lobe atrophy
(MTA) score, a global cortical atrophy (GCA) scale, identification
of Intracranial bleeds, hemorrhage, microbleeds and their volume
analysis, a fracture detection, a midline shift identification, a
measurement of the midline shift identification and the at least
one structure with respect to the midline shift identification,
identifying a pathology associated with the at least one structure,
classifying the pathology identified, a tissue density
identification, an infarct identification, a Penumbra-core-viable
tissue identification, classification and volume calculation,
diffusion-weighted imaging (DWI) maps and apparent diffusion
coefficient (ADC) maps of the at least one structure, perfusion
maps comprising resting state functional magnetic resonance imaging
(rsfMRI), an alberta stroke programme early CT score (ASPECTS)
calculation, a collateral detection, a mismatch ratio calculation,
an angiography labeling and/or annotation, a large vessel occlusion
(LVO) detection, an Hypoperfusion index calculation, Diffusion
tensor imaging (DTI) fiber tracks, neural pathway connectivity
maps, correlation between a signal input, an image input and the
text input, classifying the signal input, identifying a normal
signal, identifying an abnormal signal, identifying a pre-ictal
signal, identifying an ictal signal, extracting symptoms, and
grading of condition specific effects.
In yet another embodiment, the processor comprises a graphical
processing unit (GPU).
In yet another aspect, a non-transitory computer storage medium
storing a sequence of instructions, which when executed by a
processor, causes: obtaining one or more first images of a region
of interest of an anatomy from an image source, obtaining at least
one of a text input, and one or more physiological signals of a
patient, automatically segmenting, through a neural network, one or
more second images of at least one structure that resides within
the one or more first images, extracting one or more volumes of the
at least one structure from the one or more first images of the
region of interest, determining a feature associated with the at
least one structure based on the one or more volumes and one or
more inputs; and rendering the feature in at least one of a
two-dimensional (2D) format, a three-dimensional (3D) format, and
at least one anatomical plane. The text input comprises information
of at least one of micro-ethnicity information, an age, a race, a
gender, a medical condition, a symptom, clinical history, a patient
history, a medical test, medication information, and a cognitive
analysis report. The feature comprises at least one of the one or
more volumes of the region of interest (ROI), a cortical thickness,
an atrophy percentage, an asymmetry index score, a subfield
volumetry of the region of interest, annular volume changes, a
progressive supranuclear palsy (psp) index score, a magnetic
resonance perfusion imaging (MRPI) score, a frontal horn width to
intercaudate distance ratio (FH/CC), a medial temporal lobe atrophy
(MTA) score, a global cortical atrophy (GCA) scale, identification
of Intracranial bleeds, hemorrhage, microbleeds and their volume
analysis, a fracture detection, a midline shift identification, a
measurement of the midline shift identification and the at least
one structure with respect to the midline shift identification,
identifying a pathology associated with the at least one structure,
classifying the pathology identified, a tissue density
identification, an infarct identification, a Penumbra-core-viable
tissue identification, classification and volume calculation,
diffusion-weighted imaging (DWI) maps and apparent diffusion
coefficient (ADC) maps of the at least one structure, perfusion
maps comprising resting state functional magnetic resonance imaging
(rsfMRI), an alberta stroke programme early CT score (ASPECTS)
calculation, a collateral detection, a mismatch ratio calculation,
an angiography labeling and/or annotation, a large vessel occlusion
(LVO) detection, an Hypoperfusion index calculation, Diffusion
tensor imaging (DTI) fiber tracks, neural pathway connectivity
maps, correlation between a signal input, an image input and the
text input, classifying the signal input, identifying a normal
signal, identifying an abnormal signal, identifying a pre-ictal
signal, identifying an ictal signal, extracting symptoms, and
grading of condition specific effects.
In yet another aspect, a non-transitory computer storage medium
storing a sequence of instructions, which when executed by a
processor, causes: obtaining one or more first images of a region
of interest of an anatomy from an image source, obtaining at least
one of a text input, and one or more physiological signals of a
patient, automatically segmenting one or more second images of at
least one structure that resides within the one or more first
images, estimating one or more quantitative volumes of the at least
one structure, and predicting a prognosis based on comparison of
the one or more quantitative volumes of the at least one structure
with one or more reference volumes, the text input, and the one or
more physiological signals. The text input comprises information of
at least one of micro-ethnicity information, an age, a race, a
gender, a medical condition, a symptom, clinical history, a patient
history, a medical test, medication information, and a cognitive
analysis report.
In yet another aspect, a non-transitory computer storage medium
storing a sequence of instructions, which when executed by a
processor, causes: obtaining one or more first images of a region
of interest of an anatomy from an image source, obtaining a text
input, automatically segmenting, through a neural network, one or
more second images of at least one structure that resides within
the one or more first images, extracting one or more volumes of the
at least one structure from the one or more first images of the
region of interest, determining a feature associated with the at
least one structure based on the one or more volumes and one or
more inputs; and rendering the feature in at least one of a
two-dimensional (2D) format, a three-dimensional (3D) format, and
at least one anatomical plane. The text input comprises
micro-ethnicity information of a patient.
In one embodiment, the method further comprises: obtaining one or
more physiological signals of the patient from a signal source.
In another embodiment, the method further comprises detecting at
least one spike within the one or more physiological signals that
indicates abnormality, and predicting a prognosis based on
correlation and integrated analysis of the at least one spike
detected and the one or more volumes.
In yet another embodiment, the method further comprises detecting
an abnormal region, using the neural network, in the one or more
volumes based on the at least one spike detected, and indicating
the abnormal region using a different identifier.
In yet another embodiment, the method further comprises detecting
an abnormal region, using the neural network, in the one or more
physiological signals based on volumetric analysis, and indicating
the abnormal region comprising the at least one spike, using a
different identifier.
In yet another embodiment, the method further comprises
automatically transforming the one or more physiological signals
from a sinusoidal wave format to a quantitative format, and
predicting a prognosis based on correlation and integrated analysis
of at least one spike detected and the numerical representation of
the one or more physiological signals. The quantitative format
comprises a numerical representation of the one or more
physiological signals.
In yet another aspect, a method is described herein. The method
comprises: obtaining one or more physiological signals of a patient
from a signal source, obtaining a text input of the patient,
automatically detecting, using artificial intelligence, at least
one spike within the one or more physiological signals that
indicates abnormality, and predicting a prognosis based on the at
least one spike detected from the one or more physiological
signals, and the micro-ethnicity information. The text input
comprises micro-ethnicity information.
In one embodiment, the method further comprises: generating an
analysis report, based on the prognosis, comprising a snippet, and
a graphical representation of the prognosis.
In another embodiment, the method further comprises indicating a
portion of the one or more physiological signals where the at least
one spike, indicating the abnormality, is located.
In yet another embodiment, the signal source comprises one of (a) a
physiological signal acquisition unit, and (b) a computing
unit.
In yet another embodiment, the method further comprises
pre-processing the one or more physiological signals. The
pre-processing comprises at least one of: filtering one or more
noises associated with the one or more physiological signals, and
removing artifacts associated with the one or more physiological
signals.
In yet another embodiment, filtering the one or more noises
associated with the one or more physiological signals comprises:
passing the one or more physiological signals through at least one
of a notch filter, and a bandpass filter.
In yet another embodiment, the method further comprises
post-processing the one or more physiological signals using
artificial intelligence. The post-processing comprises: comparing
one or more first physiological signals obtained at a first
instance and one or more second physiological signals obtained at a
second instance, predicting the prognosis based on comparison of
the one or more first physiological signals and the one or more
second physiological signals, estimating one of a progression and a
regression of the prognosis associated with the patient between the
first instance and the second instance, and generating an analysis
report, based on the prognosis, comprising a snippet, and a
graphical representation of the prognosis.
In yet another embodiment, the method further comprises: obtaining
one or more first images of an anatomy of the patient from an image
source, automatically segmenting, through a neural network, one or
more second images of at least one structure that resides within
the one or more first images, extracting one or more volumes of the
at least one structure from the one or more first images,
determining a feature associated with the at least one structure
based on the one or more volumes and one or more inputs; and
rendering the feature in at least one of a two-dimensional (2D)
format, a three-dimensional (3D) format, and at least one
anatomical plane.
In yet another embodiment, the method further comprises overlaying
the one or more physiological signals as a heat map on the one or
more volumes of the at least one structure, and predicting an
orientation, a position, a shape, and a source of at least one
abnormality within the at least one structure.
In yet another embodiment, the text input further comprises at
least one of a cognitive score, a patient history, and clinical
information.
In yet another embodiment, the method comprises: predicting the
prognosis based on the at least one spike detected from the one or
more physiological signals, the micro-ethnicity information and at
least one of the cognitive score, the patient history, and the
clinical information.
In yet another aspect, a method is described herein. The method
comprises: obtaining one or more physiological signals of a patient
from a signal source, obtaining a text input of the patient,
automatically transforming the one or more physiological signals
from a sinusoidal wave format to a quantitative format, predicting
a prognosis, using an artificial intelligence, based on the
numerical representation of the one or more physiological signals,
and generating an analysis report, based on the prognosis,
comprising a snippet, and a graphical representation of the
prognosis. The quantitative format comprises a numerical
representation of the one or more physiological signals. The text
input comprises micro-ethnicity information.
In yet another aspect, a method is described herein. The method
comprises: obtaining at least one of one or more physiological
signals of a patient in response to at least one stimulus applied
to the patient, obtaining a text input of the patient, predicting
at least one of cognitive performance and cognitive deficits, using
an artificial intelligence, of the patient based on the one or more
physiological signals, and the micro-ethnicity information, and
generating an analysis report, based on at least one of the
cognitive performance, and the cognitive deficits, comprising a
snippet, and a graphical representation of a prognosis. The at
least one stimulus comprises at least one of a tangible stimulus,
and an intangible stimulus. The text input comprises
micro-ethnicity information.
In an embodiment, the at least one stimulus comprises an auditory
stimulus, a visual stimulus, an olfactory stimulus, and a palpable
stimulus.
In yet another aspect, a system is described herein. The system
comprises a server. The server comprises a memory, and a processor
communicatively coupled to the memory. The processor is operable
to: obtain one or more physiological signals of a patient from a
signal source, obtain a text input of the patient, automatically
detect, using artificial intelligence, at least one spike within
the one or more physiological signals that indicates abnormality,
and predict a prognosis based on the at least one spike detected
from the one or more physiological signals, and the micro-ethnicity
information. The text input comprises micro-ethnicity
information.
In another embodiment, a system is described herein. The server
comprises a memory, and a processor communicatively coupled to the
memory. The processor operable is to obtain one or more
physiological signals of a patient from a signal source, obtain a
text input of the patient, automatically transform the one or more
physiological signals from a sinusoidal wave format to a
quantitative format, predict a prognosis, using an artificial
intelligence, based on the numerical representation of the one or
more physiological signals, and generate an analysis report, based
on the prognosis, comprising a snippet, and a graphical
representation of the prognosis. The text input comprises
micro-ethnicity information. The quantitative format comprises a
numerical representation of the one or more physiological
signals.
In yet another aspect, a system is described herein. The system
comprises a server comprising a memory, and a processor
communicatively coupled to the memory. The processor is operable
to: obtain one or more physiological signals of a patient in
response to at least one stimulus applied to the patient, obtain a
text input of the patient, predict at least one of cognitive
performance and cognitive deficits, using an artificial
intelligence, of the patient based on the one or more physiological
signals, and the micro-ethnicity information, and generate an
analysis report, based on at least one of the cognitive
performance, and the cognitive deficits, comprising a snippet, and
a graphical representation of a prognosis. The at least one
stimulus comprises at least one of a tangible stimulus, and an
intangible stimulus. The text input comprises micro-ethnicity
information.
In yet another aspect, a non-transitory computer storage medium is
described herein. The non-transitory computer storage medium stores
a sequence of instructions, which when executed by a processor,
causes: obtaining one or more physiological signals of a patient
from a signal source, obtaining a text input of the patient,
automatically detecting, using artificial intelligence, at least
one spike within the one or more physiological signals that
indicates abnormality, and predicting a prognosis based on the at
least one spike detected from the one or more physiological
signals, and the micro-ethnicity information. The text input
comprises micro-ethnicity information.
In yet another aspect, a non-transitory computer storage medium is
described herein. The non-transitory computer storage medium
storing a sequence of instructions, which when executed by a
processor, causes: obtaining one or more physiological signals of a
patient from a signal source, obtaining a text input of the
patient, automatically transforming the one or more physiological
signals from a sinusoidal wave format to a quantitative format,
predicting a prognosis, using an artificial intelligence, based on
the numerical representation of the one or more physiological
signals, and generating an analysis report based on the prognosis.
The analysis report comprising at least one of a snippet, and a
graphical representation of the prognosis. The quantitative format
comprises a numerical representation of the one or more
physiological signals. The text input comprises micro-ethnicity
information.
In yet another aspect, a non-transitory computer storage medium is
described herein. The non-transitory computer storage medium
storing a sequence of instructions, which when executed by a
processor, causes: obtaining at least one of one or more
physiological signals of a patient in response to at least one
stimulus applied to the patient, obtaining a text input of the
patient, predicting at least one of cognitive performance and
cognitive deficits, using an artificial intelligence, of the
patient based on the one or more physiological signals, and the
micro-ethnicity information, and generating an analysis report
based on at least one of the cognitive performance, and the
cognitive deficits. The analysis report comprising at least one of
a snippet, and a graphical representation of a prognosis. The text
input comprises micro-ethnicity information. The at least one
stimulus comprises at least one of a tangible stimulus, and an
intangible stimulus.
In yet another aspect, a method is described. The method comprises:
obtaining one or more first images of a region of interest of an
anatomy from an image source; obtaining at least one of a text
input, and one or more physiological signals of a patient, wherein
the text input comprises information of at least one of
micro-ethnicity information, an age, a race, a gender, a medical
condition, a symptom, clinical history, a patient history, a
medical test, medication information, and a cognitive analysis
report; automatically segmenting, through a neural network, one or
more second images of at least one structure that resides within
the one or more first images; extracting one or more volumes of the
at least one structure from the one or more first images of the
region of interest; determining a feature associated with the at
least one structure based on the one or more volumes and one or
more inputs; and rendering the feature in at least one of a
two-dimensional (2D) format, a three-dimensional (3D) format, and
at least one anatomical plane.
In an embodiment, the method further comprises: performing a second
quality analysis manually on the one or more second images that are
segmented; and determining whether the one or more second images,
that are segmented, passes the second quality analysis.
In another embodiment, the method further comprises: providing a
user interface to the user when the one or more second images that
are segmented fails the second quality analysis; manually editing
and correcting at least one of boundaries and the one or more
volumes of the at least one structure based on one or more inputs
received from the user; and creating a mask for the at least one
structure.
In yet another embodiment, the method further comprises: training
the neural network using at least one of (a) the one or more first
images, (b) the information of at least one of the micro-ethnicity
information, the age, the race, the gender, the medical condition,
the symptom, the clinical history, the patient history, the medical
test, the medication information, and the cognitive analysis
report, (c) the one or more physiological signals, (d) the one or
more volumes of the at least one structure, (e) one or more
reference volumes, and (f) one or more reference segmented second
images.
In yet another embodiment, the method further comprises: creating a
log for the mask using the one or more inputs received from the
user; retraining the neural network based on the log created; and
automatically segmenting, through the neural network, the one or
more second images of the at least one structure in future based on
the retraining provided to the neural network.
In yet another embodiment, the method further comprises: assigning
a voxel of a mask of the one or more second images, that are
segmented, as a unit; tabulating a plurality of units in the mask;
and estimating one or more quantitative volumes of the at least one
structure from the plurality of units.
In yet another embodiment, the method further comprises: recording
the one or more volumes of the at least one structure in a
database; and categorizing the one or more volumes of the at least
one structure in the database with respect to one or more
categories of at least one of the micro-ethnicity information, an
intracranial volume (ICV), the age, the race, the gender, a family
history, the clinical history, the patient history, the symptom,
psych analysis information, brain dominance information, food
habitat information, stress information, and the medical
condition.
In yet another embodiment, the method further comprises: extracting
one or more boundaries of the at least one structure from the one
or more first images; and populating one or more voxels within the
one or more boundaries of the at least one structure using one or
more identifiers.
In yet another embodiment, the method further comprises: performing
at least one of adding, and deleting one or more voxels within the
boundaries of the at least one structure based on the one or more
inputs received from the user.
In yet another embodiment, the method further comprises: detecting
at least one spike within the one or more physiological signals
that indicates abnormality; and predicting a prognosis based on
correlation and integrated analysis of the at least one spike
detected, the text input, and the one or more volumes.
In yet another embodiment, the method further comprises:
correlating with at least one of temporal resolution and spatial
resolution of an image input and detecting an abnormal region,
using the neural network, in the one or more volumes based on the
at least one spike detected; and indicating the abnormal region
using a different identifier.
In yet another embodiment, the method further comprises
transforming automatically the one or more physiological signals
from a sinusoidal wave format to a quantitative format; predicting
a prognosis based on correlation with an image input and integrated
analysis of at least one spike detected and the numerical
representation of the one or more physiological signals; and
generating an analysis report based on the prognosis. The analysis
report comprising at least one of a feature, a snippet describing
the prognosis, one or more volumes of the at least one structure,
one or more quantitative volumes, and a graphical representation of
the prognosis. The quantitative format comprises a numerical
representation of the one or more physiological signals.
In yet another embodiment, the one or more reference volumes range
between 25th and 95th percentile. The 25th and the 95th percentile
are calculated by matching at least one of the age, the gender, the
micro-ethnicity information, and an intracranial volume (ICV) of
the patient with a normative population of individuals and then
deriving the 25th and the 95th percentile references.
In yet another embodiment, the 25th and the 95th percentile is
calculated by matching the medical condition of the patient with a
population of individuals having the medical condition and then
deriving the 25th and the 95th percentile.
In yet another embodiment, the method further comprises: predicting
a first prognosis state of the patient based at least one of the
medical condition, and first medication information of the patient
at a first point of time and generating a first analysis report;
and predicting a second prognosis state of the patient based on at
least one of the medical condition, and second medication
information of the patient at a second point of time and generating
a second analysis report.
In yet another embodiment, the method further comprises: comparing
the first prognosis state and the second prognosis state;
determining a percentage of one of a deterioration and an
improvement in at least one of one or more volumes, and the one or
more quantitative volumes based on the comparison of the first
prognosis state and the second prognosis state; and training, a
neural network, using at least one of medical condition, the first
medication information, the second medication information, and the
percentage of the deterioration or the improvement in at least one
of one or more volumes, and the one or more quantitative volumes at
a plurality of different point of times.
In yet another embodiment, the method further comprises: detecting
a diagnosis, via the neural network, at a third point of time by
comparing the first prognosis state and the second prognosis state
based on the training; performing a predictive prognosis, via the
neural network, and predicting a third prognosis state of the
patient at the third point of time based on the training; and
performing a predictive prognosis, via the neural network, and
predicting a third prognosis state of the patient at the third
point of time based on the training; and generating a third
analysis report comprising a clinical analytical outcome at the
third point of time.
The server described herein provides clinicians objective analysis
to aid in their assessment of a patient's prognosis. The server
further supports a physician's clinical impression with
quantitative numbers. The server further performs on-going multi
time point evaluation to track structural volumetric changes over
time. The server presents earlier insights about accelerated
neurodegeneration which assists physicians in identifying, treating
and lifestyle planning for such patients. The server helps as a
Neuro-Imaging Tool in conduct of Clinical Trials to determine
eligibility/monitor progress and as a Clinical End Point for multi
centric Neurology Clinical Trials. The server assists in clinical
research in studying population and disease characteristics. The
server provides services such as a Neuro-Imaging tool for Medical
Devices Companies developing products for Imaging. The server
further assists in acute cases like stroke and traumatic brain
injury (TBI) can be escalated as fast as possible. The server
further assists in performing volumetric analysis remotely. The
server further assists in early prognosis for Neurodegeneration.
The server further assists in Development of reference ranges for
the Indian Population. The server further assists in connecting
hospitals and diagnostic centers to doctors in the urban areas. The
structure-based analysis report adds objectivity to the physician's
report. The server is useful in determining the physiological age
of the brain, allowing to know the state of his/her general brain
health.
In an embodiment, the server can detect stroke, haemorrhage region,
haemorrhage types (intraparenchymal, sub dural, extradural,
subarachnoid, intraventricular), Segmentation of haemorrhage, Total
haemorrhage volume, measurement of oedema/oedematous tissue using
HU values, Measurement of midline shift and Lobe herniation;
Detection of Skull and cervical fractures; and Spinal cord
evaluation (Atlanto-axial). The server is also capable of
performing Segmentation of bleed, Fracture detection, Measurement
of midline shift, Region of bleed, Differentiation between bleed,
calcification and bone and measurement of HU (Hounsfield unit)
value, differentiation between normal tissue density and
oedematous, extracting Intra and Extra-ventricular volume; and
extracting Superior and Inferior tentorium CSF volume. The server
is also capable of epilepsy, memory dementia, pre-Alzheimer's
diagnostics, etc.
The server provides the following aspects on various MRI sequences.
Conventional MR images may not show positive findings in cases of
ischemic infarction for 8 to 12 hours after onset, a time period
beyond that when neuroprotective drugs are most likely to be given
and more likely to be effective. Diffusion weighted MR images, on
the other hand, can show regions of ischemic injury within minutes
after stroke onset. The server performs comparison between ADC and
perfusion imaging to understand blood flows. The server is also
capable of overlaying the diffusion map onto a T1 map for better
understanding of structural mapping. The server also easily
interprets magnetic resonance angiography (MRA) maps and identifies
brain regions with reduced blood vessel density. The server then
performs comparison between Apparent diffusion coefficient (ADC)
and perfusion imaging and overlay diffusion map onto the T1 map
(e.g., Diffusion-weighted imaging (DWI)) for better understanding
of structural mapping.
The methods and systems disclosed herein may be implemented in any
means for achieving various aspects and may be executed in a form
of a non-transitory machine-readable medium embodying a set of
instructions that, when executed by a machine, causes the machine
to perform any of the operations disclosed herein. Other aspects
will be apparent from the accompanying drawings and from the
detailed description that follows.
BRIEF DESCRIPTION OF THE DRAWINGS
The patent or application file contains at least one drawing
executed in color. Copies of this patent or patent application
publication with color drawing(s) will be provided by the Office
upon request and payment of the necessary fee.
Color drawings have been submitted in this application because in
figures such as FIG. 11d, FIG. 11e, FIG. 12e, FIG. 12f, FIG. 12g,
FIG. 12h, FIG. 12i, FIG. 12j, FIG. 12k, FIG. 13d, FIG. 13e, FIG.
13f, FIG. 13g, FIG. 13h, FIG. 14c, FIG. 15a, FIG. 15b, FIG. 15c,
FIG. 15d, FIG. 16a, FIG. 16b, FIG. 16c, FIG. 17a, FIG. 17c, FIG.
17d, FIG. 17e, FIG. 17f, FIG. 17g, FIG. 17h, FIG. 18, FIG. 19, FIG.
20d, FIG. 20e, FIG. 21d, FIG. 21e, FIG. 21f, FIG. 21g, FIG. 22c,
FIG. 22d, FIG. 22e, FIG. 22f, FIG. 23c, FIG. 24c, FIG. 24d, FIG.
24e, FIG. 24f, FIG. 24g, FIG. 24h, FIG. 24i, FIG. 24j, FIG. 25e,
FIG. 25f, FIG. 25g, FIG. 25h, FIG. 25i, FIG. 26c, FIG. 26d, FIG.
26e, FIG. 26f, FIG. 26g, FIG. 26h, FIG. 27c, FIG. 27d, FIG. 27e,
FIG. 27f, FIG. 27g, FIG. 28b, FIG. 29a, FIG. 29b, FIG. 29c, FIG.
32c, FIG. 38a, FIG. 38b, FIG. 38c, FIG. 38d, and FIG. 38e,
different colors represent different structures, different
boundaries, different volumes, and variations in different
graphical representations. The variation in color gives obvious
visual cues about how the phenomenon is clustered or varies. The
colors in the abovementioned figures have specific meaning and
denote specific structures in the region of interest in a
standardized way. For example, in FIG. 11e, the blue color
indicates the region of interest in a left side, and the pink color
indicates the region of interest in a right side. Further, the
colors in the abovementioned drawings are used for easy
interpretation of output to the users in a standardized way. The
usage of colors is the only way to distinguish and delineate the
structures within the region of interest exactly. If shadings are
used to distinguish and delineate the structures, the structures
cannot be exactly portrayed or displayed for the user's assessment.
The corners or edges of the region of interest may be hidden due to
shading, and therefore, might result in possibility of a medical
error. Thus, the color drawing is the only practical medium by
which aspects of the claimed subject matter may be accurately
conveyed.
In the present disclosure, reference is made to the accompanying
drawings, which form a part hereof. In the drawings, similar
symbols typically identify similar components, unless context
dictates otherwise. Various embodiments described in the detailed
description, and drawings, are illustrative and not meant to be
limiting. Other embodiments may be used, and other changes may be
made, without departing from the spirit or scope of the subject
matter presented herein. It will be understood that the aspects of
the present disclosure, as generally described herein, and
illustrated in the Figures, can be arranged, substituted, combined,
separated, and designed in a wide variety of different
configurations, all of which are contemplated herein. The
embodiments are illustrated by way of example and not limitation in
the figures of the accompanying drawings, in which like references
indicate similar elements and in which:
FIG. 1 illustrates a schematic view of a system, according to one
or more embodiments.
FIG. 2 illustrates an exploded view of a server, according to one
or more embodiments.
FIG. 3 illustrates an overview of a system, according to one or
more embodiments.
FIG. 4 illustrates a multivariate pattern analysis performed by a
system, according to one or more embodiments.
FIG. 5 illustrates a method of structure-based analysis report
generation, according to one or more embodiments.
FIG. 6 illustrates users of a system, according to one or more
embodiments.
FIG. 7 illustrates a process flow of a system, according to one or
more embodiments.
FIGS. 8 and 9 illustrate a system architecture, according to one or
more embodiments.
FIG. 10 illustrates a workflow, according to one or more
embodiments.
FIG. 11a-11e illustrate a process of segmentation of Hippocampus,
according to one or more embodiments.
FIG. 12a-12k illustrate a process of segmentation of Ventricles,
according to one or more embodiments.
FIG. 13a-13h illustrate a process of segmentation of a Whole Brain,
according to one or more embodiments.
FIG. 14a-14c illustrate a process of segmentation of an
intracranial volume (ICV), according to one or more
embodiments.
FIG. 15a-15d illustrate a process of segmentation of Cerebrum,
according to one or more embodiments.
FIG. 16a-16c illustrate a process of segmentation of Cerebellum,
according to one or more embodiments.
FIG. 17a-17h illustrate a process of segmentation of Brainstem,
according to one or more embodiments.
FIG. 18 illustrates a process of segmentation of Midbrain,
according to one or more embodiments.
FIG. 19 illustrates a process of segmentation of Pons, according to
one or more embodiments.
FIG. 20a-20e illustrate a process of segmentation of Amygdala,
according to one or more embodiments.
FIG. 21a-21g illustrate a process of segmentation of Basal Ganglia,
according to one or more embodiments.
FIG. 22a-22f illustrate a process of segmentation of Thalamus,
according to one or more embodiments.
FIG. 23a-23c illustrate a process of segmentation of Substantia
Nigra, according to one or more embodiments.
FIG. 24a-24j illustrate a process of segmentation of Frontal Lobes,
according to one or more embodiments.
FIG. 25a-25i illustrate a process of segmentation of Parietal
Lobes, according to one or more embodiments.
FIG. 26a-26h illustrate a process of segmentation of Occipital
Lobes, according to one or more embodiments.
FIG. 27a-27g illustrate a process of segmentation of Temporal
Lobes, according to one or more embodiments.
FIG. 28a and 28b illustrate a structure-based analysis report,
according to one or more embodiments.
FIG. 29a-29c illustrate an integrated analysis report showing an
integrated multimodal analysis of an image input, a text input, and
a signal input, according to one or more embodiments.
FIG. 30a-30b illustrate an EEG detailed report, according to one or
more embodiments.
FIG. 31 illustrates monitoring of one or more physiological
signals, according to one or more embodiments.
FIG. 32a illustrates a screenshot of a user interface that allows a
user to upload patient details, according to one or more
embodiments.
FIG. 32b illustrates a screenshot of a user interface that allows a
user to view patient details, according to one or more
embodiments.
FIG. 32c illustrates a screenshot of a user interface rendering a
segmented image, according to one or more embodiments.
FIG. 32d illustrates a screenshot of a user interface that allows a
user to view patient details, according to one or more
embodiments.
FIG. 33 illustrates the processing of EEG signals, according to one
or more embodiments.
FIG. 34 illustrates a data flow of a system, according to one or
more embodiments.
FIG. 35 illustrates a workflow of a system, according to one or
more embodiments.
FIG. 36 further illustrates an architecture of a system, according
to one or more embodiments.
FIG. 37 illustrates an architecture of a system, according to one
or more embodiments.
FIG. 38a-38e illustrate an analysis report generated based on
condition specific analysis, according to one or more
embodiments.
Other aspects of the present embodiments will be apparent from the
accompanying drawings and from the detailed description that
follows.
DETAILED DESCRIPTION
Although the following detailed description contains many specifics
for the purpose of illustration, a person of ordinary skill in the
art will appreciate that many variations and alterations to the
following details can be made and are considered to be included
herein.
Accordingly, the following embodiments are set forth without any
loss of generality to, and without imposing limitations upon, any
claims set forth. It is also to be understood that the terminology
used herein is for the purpose of describing particular embodiments
only, and is not intended to be limiting. Unless defined otherwise,
all technical and scientific terms used herein have the same
meaning as commonly understood by one of ordinary skill in the art
to which this disclosure belongs.
The articles "a" and "an" are used herein refers to one or to more
than one (i.e., to at least one) of the grammatical object of the
article. By way of example, "an element" means one element or more
than one element.
No element, act, or instruction used herein should be construed as
critical or essential unless explicitly described as such. Also, as
used herein, the articles "a" and "an" are intended to include
items, and may be used interchangeably with "one or more."
Furthermore, as used herein, the term "set" is intended to include
items (e.g., related items, unrelated items, a combination of
related items, and unrelated items, etc.), and may be used
interchangeably with "one or more." Where only one item is
intended, the term "one" or similar language is used. Also, as used
herein, the terms "has," "have," "having," or the like are intended
to be open-ended terms. Further, the phrase "based on" is intended
to mean "based, at least in part, on" unless explicitly stated
otherwise.
The terms "first," "second," "third," "fourth," and the like in the
description and in the claims, if any, are used for distinguishing
between similar elements and not necessarily for describing a
particular sequential or chronological order. It is to be
understood that the terms so used are interchangeable under
appropriate circumstances such that the embodiments described
herein are, for example, capable of operation in sequences other
than those illustrated or otherwise described herein. Furthermore,
the terms "include," and "have," and any variations thereof, are
intended to cover a non-exclusive inclusion, such that a process,
method, system, article, device, or apparatus that comprises a list
of elements is not necessarily limited to those elements, but may
include other elements not expressly listed or inherent to such
process, method, system, article, device, or apparatus.
The terms "left," "right," "front," "back," "top," "bottom,"
"over," "under," and the like in the description and in the claims,
if any, are used for descriptive purposes and not necessarily for
describing permanent relative positions. It is to be understood
that the terms so used are interchangeable under appropriate
circumstances such that the embodiments of the apparatus, methods,
and/or articles of manufacture described herein are, for example,
capable of operation in other orientations than those illustrated
or otherwise described herein.
In this disclosure, "comprises," "comprising," "containing" and
"having" and the like can have the meaning ascribed to them in U.S.
Patent law and can mean "includes," "including," and the like, and
are generally interpreted to be open ended terms. The terms
"consisting of" or "consists of" are closed terms, and include only
the components, structures, steps, or the like specifically listed
in conjunction with such terms, as well as that which is in
accordance with U.S. Patent law. "Consisting essentially of" or
"consists essentially of" have the meaning generally ascribed to
them by U.S. Patent law. In particular, such terms are generally
closed terms, with the exception of allowing inclusion of
additional items, materials, components, steps, or elements, that
do not materially affect the basic and novel characteristics or
function of the item(s) used in connection therewith. For example,
trace elements present in a composition, but not affecting the
composition's nature or characteristics would be permissible if
present under the "consisting essentially of" language, even though
not expressly recited in a list of items following such
terminology. When using an open-ended term in this written
description, like "comprising" or "including," it is understood
that direct support should also be afforded to "consisting
essentially of" language as well as "consisting of" language as if
stated explicitly and vice versa.
As used herein, the term "about" is used to provide flexibility to
a numerical range endpoint by providing that a given value may be
"a little above" or "a little below" the endpoint. However, it is
to be understood that even when the term "about" is used in the
present specification in connection with a specific numerical
value, that support for the exact numerical value recited apart
from the "about" terminology is also provided.
Reference throughout this specification to "an example", "an
instance", "for example" means that a particular aspect, structure,
or characteristic described in connection with the example is
included in at least one embodiment. Thus, appearances of the
phrases "in an example" in various places throughout this
specification are not necessarily all referring to the same
embodiment.
Implementations and all of the functional operations described in
this specification may be realized in digital electronic circuitry,
or in computer software, firmware, or hardware, including the
structures disclosed in this specification and their structural
equivalents, or in combinations of one or more of them.
Implementations may be realized as one or more computer program
products, i.e., one or more modules of computer program
instructions encoded on a computer readable medium for execution
by, or to control the operation of, data processing apparatus. The
computer readable medium may be a machine-readable storage device,
a machine-readable storage substrate, a memory device, a
composition of matter affecting a machine-readable propagated
signal, or a combination of one or more of them. The term
"computing system" encompasses all apparatus, devices, and machines
for processing data, including by way of example a programmable
processor, a computer, or multiple processors or computers. The
apparatus may include, in addition to hardware, code that creates
an execution environment for the computer program in question,
e.g., code that constitutes processor firmware, a protocol stack, a
database management system, an operating system, or a combination
of one or more of them. A propagated signal is an artificially
generated signal, e.g., a machine-generated electrical, optical, or
electromagnetic signal that is generated to encode information for
transmission to suitable receiver apparatus.
The actual specialized control hardware or software code used to
implement these systems and/or methods is not limited to the
implementations. Thus, the operation and behavior of the systems
and/or methods were described herein without reference to specific
software code--it being understood that software and hardware can
be designed to implement the systems and/or methods based on the
description herein.
A computer program (also known as a program, software, software
application, script, or code) may be written in any appropriate
form of programming language, including compiled or interpreted
languages, and it may be deployed in any appropriate form,
including as a standalone program or as a module, component,
subroutine, or other unit suitable for use in a computing
environment. A computer program does not necessarily correspond to
a file in a file system. A program may be stored in a portion of a
file that holds other programs or data (e.g., one or more scripts
stored in a markup language document), in a single file dedicated
to the program in question, or in multiple coordinated files (e.g.,
files that store one or more modules, sub programs, or portions of
code). A computer program may be deployed to be executed on one
computer or on multiple computers that are located at one site or
distributed across multiple sites and interconnected by a
communication network.
The processes and logic flows described in this specification may
be performed by one or more programmable processors executing one
or more computer programs to perform functions by operating on
input data and generating output. The processes and logic flows may
also be performed by, and apparatus may also be implemented as,
special purpose logic circuitry, for example without limitation, a
PLC (Programmable Logic Controller), an FPGA (field programmable
gate array), an ASIC (application specific integrated circuit),
Application-specific Standard Products (ASSPs), System-on-a-chip
systems (SOCs), Complex Programmable Logic Devices (CPLDs),
etc.
Processors suitable for the execution of a computer program
include, by way of example, both general and special purpose
microprocessors, and any one or more processors of any appropriate
kind of digital computer. Generally, a processor will receive
instructions and data from a read only memory or a random-access
memory or both. Elements of a computer can include a processor for
performing instructions and one or more memory devices for storing
instructions and data. Generally, a computer will also include, or
be operatively coupled to receive data from or transfer data to, or
both, one or more mass storage devices for storing data, e.g.,
magnetic, magneto optical disks, or optical disks. However, a
computer need not have such devices. Moreover, a computer may be
embedded in another device, e.g., a mobile telephone, a personal
digital assistant (PDA), a mobile audio player, a Global
Positioning System (GPS) receiver, to name just a few. Computer
readable media suitable for storing computer program instructions
and data include all forms of non-volatile memory, media and memory
devices, including by way of example semiconductor memory devices,
e.g., Erasable Programmable Read Only Memory (EPROM), Electrically
Erasable Programmable Read Only Memory (EEPROM), and flash memory
devices; magnetic disks, e.g., internal hard disks or removable
disks; magneto optical disks; and compact disk Read-only memory (CD
ROM) and Digital Versatile Disk-Read-only memory (DVD-ROM) disks.
The processor and the memory may be supplemented by, or
incorporated in, special purpose logic circuitry.
To provide for interaction with a user, implementations may be
realized on a computer having a display device, e.g., a CRT
(cathode ray tube) or LCD (liquid crystal display) monitor, for
displaying information to the user and a keyboard and a pointing
device, e.g., a mouse or a trackball, by which the user may provide
input to the computer. Other kinds of devices may be used to
provide for interaction with a user as well; for example, feedback
provided to the user may be any appropriate form of sensory
feedback, e.g., visual feedback, auditory feedback, or tactile
feedback; and input from the user may be received in any
appropriate form, including acoustic, speech, or tactile input.
Implementations may be realized in a computing system that includes
a back-end component, e.g., as a data server, or that includes a
middleware component, e.g., an application server, or that includes
a front-end component, e.g., a client computer having a graphical
user interface or a Web browser through which a user may interact
with an implementation, or any appropriate combination of one or
more such back end, middleware, or front-end components. The
components of the system may be interconnected by any appropriate
form or medium of digital data communication, e.g., a communication
network. Examples of communication networks include a local area
network ("LAN") and a wide area network ("WAN"), e.g., the
Internet.
The computing system may include clients and servers. A client and
server are generally remote from each other and typically interact
through a communication network. The relationship of client and
server arises by virtue of computer programs running on the
respective computers and having a client-server relationship to
each other.
Embodiments of the present disclosure may comprise or utilize a
special purpose or general-purpose computer including computer
hardware. Embodiments within the scope of the present disclosure
also include physical and other computer-readable media for
carrying or storing computer-executable instructions and/or data
structures. Such computer-readable media can be any available media
that can be accessed by a general purpose or special purpose
computer system. Computer-readable media that store
computer-executable instructions are physical storage media.
Computer-readable media that carry computer-executable instructions
are transmission media. Thus, by way of example, and not
limitation, embodiments of the disclosure can comprise at least two
distinctly different kinds of computer-readable media: physical
computer-readable storage media and transmission computer-readable
media.
Computer-executable instructions comprise, for example,
instructions and data which cause a general-purpose computer,
special purpose computer, or special purpose processing device to
perform a certain function or group of functions. The
computer-executable instructions may be, for example, binaries,
intermediate format instructions such as assembly language, or even
source code. Although the subject matter has been described in
language specific to structural aspects and/or methodological acts,
it is to be understood that the subject matter defined in the
appended claims is not necessarily limited to the described aspects
or acts described. Rather, the described aspects and acts are
disclosed as example forms of implementing the claims.
Physical computer-readable storage media includes RAM, ROM, EEPROM,
CD-ROM or other optical disk storage (such as CDs, DVDs, etc.),
magnetic disk storage or other magnetic storage devices, or any
other medium which can be used to store desired program code means
in the form of computer-executable instructions or data structures
and which can be accessed by a general purpose or special purpose
computer.
A "network" is defined as one or more data links that enable the
transport of electronic data between computer systems and/or
modules and/or other electronic devices. When information is
transferred or provided over a network, such as a 5G network, or
another communications connection (either hardwired, wireless, or a
combination of hardwired or wireless) to a computer, the computer
properly views the connection as a transmission medium.
Transmission media can include a network and/or data links which
can be used to carry data or desired program code means in the form
of computer-executable instructions or data structures and which
can be accessed by a general purpose or special purpose computer.
Combinations of the above are also included within the scope of
computer-readable media.
Further, upon reaching various computer system components, program
code means in the form of computer-executable instructions or data
structures can be transferred automatically from transmission
computer-readable media to physical computer-readable storage media
(or vice versa). For example, computer-executable instructions or
data structures received over a network or data link can be
buffered in RAM within a network interface module (e.g., a "NIC"),
and then eventually transferred to computer system RAM and/or to
less volatile computer-readable physical storage media at a
computer system. Thus, computer-readable physical storage media can
be included in computer system components that also (or even
primarily) utilize transmission media.
Computer-executable instructions comprise, for example,
instructions and data which cause a general-purpose computer,
special purpose computer, or special purpose processing device to
perform a certain function or group of functions. The
computer-executable instructions may be, for example, binaries,
intermediate format instructions such as assembly language, or even
source code. Although the subject matter has been described in
language specific to structural aspects and/or methodological acts,
it is to be understood that the subject is not necessarily limited
to the described aspects or acts described above. Rather, the
described aspects and acts are disclosed as example forms of
implementing subject matter.
While this specification contains many specifics, these should not
be construed as limitations on the scope of the disclosure or of
what may be claimed, but rather as descriptions of aspects specific
to particular implementations. Certain aspects that are described
in this specification in the context of separate implementations
may also be implemented in combination in a single implementation.
Conversely, various aspects that are described in the context of a
single implementation may also be implemented in multiple
implementations separately or in any suitable sub-combination.
Moreover, although aspects may be described above as acting in
certain combinations and even initially claimed as such, one or
more aspects from a claimed combination may in some cases be
excised from the combination, and the claimed combination may be
directed to a sub-combination or variation of a
sub-combination.
Similarly, while operations are depicted in the drawings in a
particular order, this should not be understood as requiring that
such operations be performed in the particular order shown or in
sequential order, or that all illustrated operations be performed,
to achieve desirable results. In certain circumstances,
multitasking and parallel processing may be advantageous. Moreover,
the separation of various system components in the implementations
described above should not be understood as requiring such
separation in all implementations, and it should be understood that
the described program components and systems may generally be
integrated together in a single software product or packaged into
multiple software products.
Even though particular combinations of aspects are disclosed in the
specification, these combinations are not intended to limit the
disclosure of possible implementations.
Further, the methods may be practiced by a computer system
including one or more processors and computer-readable media such
as computer memory. In particular, the computer memory may store
computer-executable instructions that when executed by one or more
processors cause various functions to be performed, such as the
acts recited in the embodiments.
An initial overview of technology embodiments is provided below,
and specific technology embodiments are then described in further
detail. This initial summary is intended to aid readers in
understanding the technology more quickly but is not intended to
identify key or essential technological aspects, nor is it intended
to limit the scope of the claimed subject matter.
The embodiments herein and the various aspects and advantageous
details thereof are explained more fully with reference to the
non-limiting embodiments that are illustrated in the accompanying
drawings and detailed in the following description.
In order to fully understand the scope of the invention, the
following terms used herein are hereby defined.
As used herein, "Image source" refers to any medical assessment
device including but not limited to electroencephalogram (EEG),
computed tomography (CT) Scan, magnetic resonance imaging (MRI),
Magnetoencephalography (MEG), Functional magnetic resonance imaging
(fMRI), positron emission tomography (PET), X Rays, ultrasound, 2D
Fluid-attenuated inversion recovery (FLAIR), 3D Magnetic resonance
angiography (MRA) and psychological assessment (PA) or any
computing device used to obtain and/or store images of an organ of
an organism.
In an embodiment, "image source" refers to different sources
including, but not limited to one or more of the following: medical
centers, large pharmaceutical companies (e.g., in association with
pre-clinical evaluations or during clinical trials), contract
research organizations (CRO) (for both pre-clinical and clinical
analysis), medical laboratories and practices (e.g., scanning
centers), hospitals, clinics, medical centers, small biotechnology
companies (e.g., in association with pre-clinical evaluations or
during clinical trials), and bio-medical research
organizations.
As used herein "Anatomy" refers to structure and internal workings
of an organism.
As used herein "Anatomically meaningful region" refers to a region
or a structure within an organism, functions individually and/or in
combination, that has an influence in predicting prognosis,
diagnosis, volumetric extraction, volumetric analysis, and atrophy
information. Anatomically meaningful region may also refer to a
region or a structure, individually and/or in combination, that
comprises distinct or unique functional characteristics.
In an embodiment, the term "anatomically meaningful region" refers
to a region formed as a result of analysing image data obtained via
photographing or scanning, and dividing it into spatial regions
that are anatomically and physiologically meaningful.
As used herein, the term "based on" is defined as dependent on.
As used herein, the term "a plurality of" is defined as
multiple.
As used herein, the term "memory" is defined as any device in which
information can be stored.
As used herein, the term "execute" is defined as run or launch.
As used herein, the term "instructions" is defined as software
program or machine executable code.
As used herein, "neural network" refers to a computational learning
system that uses a network of functions to understand and translate
a data input of one form into a desired output, usually in another
form.
In an embodiment, the term "neural network" refers to a
computational model implemented in software or hardware that mimics
the computational ability of a biological system using a large
number of interconnected artificial neurons. The neural network, in
the present disclosure, is trained to predict a prognosis, atrophy
and diagnosis based on structural analysis. The neural network is
also capable of predicting prognosis based on multimodal analysis
such as by receiving at least one of an image input, a text input,
and a signal input.
As used herein "Physiological signals" refers to signals that are
acquired from an organism such as at least one of but not limited
to an electrocardiography (ECG) signal, an electroencephalogram
(EEG) signal, an Electromyography (EMG), a galvanic skin response
(GSR), a blood pressure, event related potential (ERP), a pulse
rate, etc.
In an embodiment, the term signal "Physiological signals" refers to
signals acquired from an organism for measuring or detection of a
physiological parameter or change of such a parameter.
As used herein "Demographic information" refers to a collection of
data comprising age, race, gender, genotype and micro-ethnicity.
Further demographic information, used herein, refers to information
that is used to recognize, identify, classify, group, and/or
categorize an organism.
In an embodiment, the "Demographic information" refers to
demographic details of the patient which include, but are not
restricted to, name, age, gender, and bed location. The demographic
details refer to personal details that contribute to recognizing
the identity of a patient.
As used herein "image input" refers to inputs that are in pictorial
representation. The image inputs may be obtained by scanning at
least one organ or capturing images of at least one organ of an
organism. The organ may be an internal organ or an external organ.
The image input may be acquired or obtained or received from an
image source.
In an embodiment, the term "image input" refers to the medical
images that are received as an input to perform image segmentation
and predict a prognosis. In another embodiment, the "image input"
refers to digital data capable of producing a visual
representation. For instance, the term "image input" includes
digital images and digital video.
As used herein "text input" refers to inputs that are in written
format in any language. The text input comprises inputs in text
format that are entered into a machine and extracted from one or
more records. The text input may be obtained by a natural language
processing (NLP) technique. The NLP technique may be used to read,
decipher, understand, and make sense of the human languages in a
manner that is valuable and can assist in predictive prognosis and
diagnosis. The text input may also be converted from a first
language to a second language that is understandable by a
system.
In an embodiment, the term "text input" refers to an input obtained
by the server in a text format. The text inputs may comprise
details such as a medical condition, a gender, an age, a
micro-ethnicity, symptoms, physician details.
As used herein "signal input" refers to inputs that are in
graphical format. The signal input comprises physiological signals
that are acquired from an organism usually in a sinusoidal wave
format. The signal input may comprise at least one spike that may
represent abnormal functioning or normal functioning of an
organism. The signal input may comprise a recorded physiological
signal. The signal input may also be a physiological signal that is
acquired from the organism in real-time.
In an embodiment, the term "signal input" refers to input in the
form of signals. The signals may be any physiological signals that
are adapted for measuring or detection of a physiological parameter
or change of such a parameter.
As used herein "Metadata" refers to patient metadata, or
descriptive information about the patient (including demographics,
pharmaceuticals, diagnosis, etc.), that needs to be recorded in a
way that is interoperable. Additionally, Metadata also refers to
administrative metadata that needs to be included for the records
to be understandable outside of the context in which they were
created. Administrative metadata is data describing the electronic
medical records; Metadata can include information about the
controlled vocabularies and standards used, necessary information
to ensure patient privacy, and specifics of the electronic health
records' authenticity and creation.
In an embodiment, the term "metadata" refers to data that
represents information about user or system data and describes
attributes of actual user or system data. Further, metadata is data
that describes other data stored in the downloadable medical files
that may provide the functionality needed to manage and access the
data in medical files. The metadata may be protected from
manipulation and/or access using one or more methods of
encryption
As used herein "Intracranial volume (ICV)" refers to volume within
the cranium including the brain, meninges, and CSF. The ICV also
refers to an estimated volume of cranial cavity as outlined by
supratentorial dura mater or cerebral contour when dura is not
clearly detectable. The Intracranial volume sometimes refers to the
total intracranial volume (TIV).
In an embodiment, the term "Intracranial Volume (ICV)" is a
standard measure to correct for head size in different brain
studies and in AD related literature. The ICV measure, sometimes
referred to as total intracranial volume (TIV), refers to the
estimated volume of the cranial cavity as outlined by the
supratentorial dura mater or cerebral contour when dura is not
clearly detectable. ICV is often used in studies involved with
analysis of the cerebral structure under different imaging
modalities, such as Magnetic Resonance (MR), MR and Diffusion
Tensor Imaging (DTI), MR and Single-photon Emission Computed
Tomography (SPECT), Ultrasound and Computed Tomography (CT). ICV
consistency during aging makes it a reliable tool for correction of
head size variation across subjects in studies that rely on
morphological characteristics of the brain. ICV, along with age and
gender are reported as covariates to adjust for regression analysis
in investigating progressive neurodegenerative brain disorders,
such as Alzheimer's disease, aging and cognitive impairment.
As used herein "Quantitative volumes" refers to voxel-based
analysis of tissue characteristics such as volume, T2 and diffusion
density/concentration in an organ. Quantitative volumes further
refer to numerical representation of structure or volume, density
of the at least one organ in the organism.
As used herein "progression" refers to the forecast of the probable
outcome or course of a disease; the patient's chance of recovery.
Progression further to increase in severity and/or size of diseased
area at a later point.
In an embodiment, the term "progression" refers to evolution of the
disease over time. Further progression implies that a patient is
initially diagnosed with an early stage of the disease with
worsening at the current examination.
As used herein "regression" refers to a characteristic of diseases
such as decrease in severity and/or size without completely
disappearing. At a later point, symptoms may return.
In an embodiment, the term "regression" implies the presence of the
disease at the preceding examination with an improvement at the
current examination.
As used herein "diagnosis" refers to a process of identifying a
disease, condition, or injury from its signs and symptoms. A health
history, physical exam, and tests, such as blood tests, imaging
tests, and biopsies, may be used to perform a diagnosis.
In an embodiment, the term "diagnosis" refers to methods by which
the person skilled in the art can estimate and/or measure the
probability ("likelihood") whether a patient suffers from a given
disease or condition. In the present disclosure, "diagnosis" refers
to the use of the system to analyze the structural changes in at
least one organ and estimate the medical condition of a patient
based on structure. The term diagnosis also refers to an estimation
of detection of disease or medical condition of the patient based
on multimodal analysis using at least one of image input, text
input and signal input.
As used herein "prognosis" refers to a forecast, a prospect, a
prediction of what the future stage of disease will be, regarding a
single case. It also refers to the probability that an applied
treatment will be effective equals the probability that the
treatment will, in a beneficent way, alter the course and eventual
outcome of the disease.
In an embodiment, the term "prognosis" refers to a prediction of
the probable course and outcome of a clinical condition, a state,
or a disease of a patient. A prognosis of the patient is usually
made by evaluating factors or symptoms of a disease that are
indicative of an outcome of the disease. The skilled artisan will
understand that the term "prognosis" refers to an increased
probability that a certain course or outcome will occur; that is,
that a course or outcome is more likely to occur in a patient
exhibiting a given condition, when compared to those individuals
not exhibiting the condition. A prognosis may be expressed as the
amount of time a patient can be expected to survive. Alternatively,
a prognosis may refer to the likelihood that the disease goes into
remission or to the amount of time the disease can be expected to
remain in remission. A prognosis is often determined by examining
one or more prognostic factors or indicators or change in
structural volumes of at least one organ. The progression may also
refer to progression or regression status of the disease.
As used herein "biomarkers" refers to any substance, structure, or
process that can be measured in the body or its products and
influence or predict the incidence of outcome or disease.
In an embodiment, the term "biomarkers" is a clinical or biological
characteristic that provides information on the likely patient
health outcome. The biomarkers may be at least one of a prognostic
biomarker, and a predictive biomarker. A prognostic biomarker
provides information about the patient's overall disease outcome
and progression or regression of the disease, regardless of
therapy, whereas a predictive biomarker gives information about the
effect of a treatment.
As used herein "anonymization" refers to the process of turning
data into a form that does not identify and recognize individuals.
Anonymization breaks the link between data and a given participant
so that the participant cannot be identified, directly or
indirectly (e.g., through cross-referencing), from their data.
In an embodiment, the term "anonymization" refers to a process of
concealing patient identity before transferring data and images
outside the confidential confines of the patient care facility.
Medical images are commonly encoded and stored in a DICOM (digital
imaging and communications in medicine) format. DICOM images have a
header section that includes several fields, such as patient name,
patient identification, birth date, hospital name, date of
acquisition, techniques used for acquisition, etc. Key patient
identifiable fields, such as, but not limited to patient name and
patient ID, need to be anonymized before the images can be shared
with research facilities. Once the data is anonymized, the patient
identity is concealed such that one cannot trace or track the
source (e.g., patient identity, site identity, etc.) of the medical
data.
As used herein "reference volume" refers to volume of an organ or
body part within the organism that are empirically defined based on
known clinical cases. Reference volumes may be used to compare with
current estimated volumes of a patient based on at least age,
gender, micro-ethnicity and ICV to predict a prognosis and to
perform a diagnosis.
In an embodiment, the term "reference volume" refers to a volume of
at least one organ which is obtained by averaging one or more
volumes that are manually segmented. In one embodiment, the
reference volume serves as the basis to train the system in
automatic image segmentation.
As used herein "reference quantitative volume" refers to
quantitative volumes of an organ or body part within the organism
that are empirically defined based on known clinical cases.
Reference quantitative volumes may be used to compare with current
estimated quantitative volumes of a patient based on at least age,
gender, micro-ethnicity and ICV to predict a prognosis and to
perform diagnosis.
In an embodiment "reference quantitative volume" refers to
quantitative volumes of an organ or body part within the organism
that are used as a reference to make any decisions or actions in
further processing of image segmentation, volumetric analysis or
prognosis prediction.
As used herein "users" refers to a person who has privileges or has
permission to access the system.
In an embodiment "user" comprises one of a Radiologist, a Doctor, a
technical specialist, a manager, an administrator, an analyst,
etc.
As used herein "computing unit" refers to any personal digital
assistant unit comprising but not limited to a desktop, a Laptop, a
mobile phone, a handheld PC, a smart phone, etc.
In an embodiment, the term "computing unit" refers to a group of
physical components having close physical relationship with each
other and can be used as a basic unit for executing a task
As used herein "Segmented image" refers to the structures in the
image that have been "gathered" into anatomically meaningful
regions. The segmented image also refers to the image (i.e.,
anatomically meaningful region) that has been segmented from the
image of the region of interest of an anatomy. The segmented image
may be useful in volumetric extraction, volumetric analysis which
in turn may be helpful in predicting prognosis, diagnosis and
atrophy information.
In an embodiment, the term "Segmented image" refers to a collection
of pixels that is segmented from a medical image. The segmented
image may be an anatomically meaningful portion of the medical
image.
As used herein "image quality analysis" refers to the method of
determining whether the images of the region of interest meets
current industry standard of volumetric extraction. In an
embodiment, the image quality analysis refers to determining
whether the images obtained comprise a predefined magnetic strength
value more than 1.5 Tesla.
In an embodiment, the term "image quality analysis" refers to
analysing quality of the image and determining whether the image
captured is appropriate for image segmentation, volumetric
analysis, and volumetric extraction and other processing.
As used herein "database" refers to a set of computer readable
storage mediums associated with a set of computer executable
programs. The database stores the information related to the user
details, one or more first images of a region of interest of an
anatomy obtained from an image source, demographic information, and
one or more physiological signals acquired from a patient, one or
more volumes of at least one structure that resides within the one
or more first images with respect to micro-ethnicity information in
at least one of a three-dimensional (3d) format, and at least one
medical plane, one or more quantitative volumes of the at least one
structure of the region of interest categorized with respect to the
micro-ethnicity information, one or more structure-based analysis
report, one or more reference volumes, an index for the one or more
volumes, and the one or more quantitative volumes, user
identification data assigned to the patient, progression and a
regression state of prognosis and a health condition of the
patient. The database stores population-based volume/structure
standardization and big data gathering. The database in one aspect,
stores a neuroimaging data bank. The database in another aspect,
stores orthoimage data repository. The one or more volumes, the one
or more quantitative volumes, and the one or more reference volumes
are stored in a data structure.
In an embodiment, the "database" refers to an organized collection
of data, generally stored and recorded such that the data can be
accessed and updated electronically from a computing unit. Database
also refers to a systematic collection of data. Database further
supports electronic storage and manipulation of data.
As used herein "organ" refers to a body part of an organism. The
organ comprises at least one of one of a circulatory system, a
nervous system, a muscular system, an endocrine system, a
respiratory system, a digestive system, a urinary system, a
reproductive system, an integumentary system, an immune system, and
a skeletal system.
As used herein "Anatomical plane" refers to a hypothetical plane
used to transect the body, in order to describe the location,
position and orientation of structures. It includes but is not
limited to horizontal plane, coronal plane, sagittal plane,
parasagittal plane, transverse plane, anterior axillary line, mid
axillary line, midclavicular line, posterior axillary line and the
like.
In an embodiment "anatomical planes" refers to imaginary flat
surfaces or planes that pass through the body in the anatomical
position and are used to divide the body. These planes are
imaginary lines and can be vertical or horizontal and drawn through
an upright body.
As used herein "Segmentation" refers to a process that allows to
mainly define subsets of pixels or voxels of images, based on at
least one of structure, volume and density, of the region of
interest of an anatomy and define boundaries to form at least one
anatomically meaningful region. Once said subsets and boundaries
have been defined it is possible to transform each subset into a
virtual object therefore having its functional or semantic unit.
Segmentation can be done through at least one of automatically,
semi-automatically or manually.
In an embodiment, "Segmentation" refers to the detection of
boundaries of structures of at least one body part such as organs,
vessels, different types of tissue, pathologies, medical devices,
etc., in medical images of a patient. Segmentation also involves
marking the boundaries and determining the quantitative volumes of
Automatic segmentation of anatomical objects is a
prerequisite/mandatory for many medical image analysis tasks, such
as prognosis, disease diagnosis, atrophy determination, and
quantification. Segmentation is a computer aided process which
automatically segments anatomical meaningful regions (e.g., organs)
once training is provided to the system.
As used herein "Micro-ethnicity information" refers to the
information related to groups of peoples belonging to a particular
geographical area (e.g., a sub region, a sub zone, a region, a
zone, a district, a city, a state, a country, a continent etc.) who
have certain racial, cultural, religious, or other traits in
common. Patients belonging to a particular micro-ethnicity may have
unique anatomical characteristics such as volume, weight, cross
sectional area and dimensions of the internal organs (e.g.
cardiovascular organs, neural organs, orthopaedic organs, etc.), an
intracranial volume (ICV), information about previous and present
diseases, psych analysis information, brain dominance information,
cognitive measures, stress information, food habits and physical
activity habits, blood type, cholesterol level, handedness, and
comorbidity conditions, etc.
In an embodiment, the term "micro-ethnicity information" refers to
the information related to small groups and localized ethnic
communities-cum-sociological groups (`micro-ethnic groups`), or
sets of neighboring or previously neighboring groups or not
neighboring sharing common identity, and sometimes, but not always,
common origins.
As used herein "atrophy" refers to progressive degeneration or
shrinkage of at least one organ. Atrophy may help to perform
patient-specific predictive prognosis and diagnosis.
In an embodiment, the term "atrophy" refers to progressive loss of
muscle mass and/or progressive weakening and degeneration of
muscles, including skeletal and voluntary muscles, cardiac muscles
that control the heart (cardiomyopathies), and smooth muscles.
Atrophy also refers to a decrease in size of a body part, cell,
organ, or other tissue. The term "atrophy" implies that the
atrophied part was of a size normal for the individual, considering
age and circumstance, prior to the diminution.
As used herein "site" refers to a place where image segmentation,
volumetric analysis, volumetric extraction is performed. The site
also refers to a place where predictive prognosis, diagnosis, and
atrophy information is needed. The site comprises one of a
diagnostic center, a hospital, a clinic, a healthcare unit, an
organization where at least one of research, analysis, and
generating three-dimensional models of structure of organs is
performed, and the like. The site also refers to a place where
accuracy, quality of the segmented image is verified and enhanced
in terms of boundaries, volumes, shape, density, orientation,
intensity and the like.
In an embodiment, the term "site" refers to but not limited to one
or more of the following: medical centers, large pharmaceutical
companies (e.g., in association with pre-clinical evaluations or
during clinical trials), contract research organizations (CRO) (for
both pre-clinical and clinical analyzes), medical laboratories and
practices (e.g., scanning centers), hospitals, clinics, medical
centers, medical image processing organizations, Research Centers,
small biotechnology companies (e.g., in association with
pre-clinical evaluations or during clinical trials), and
bio-medical research organizations.
As used herein "volumetric extraction" refers to a process of
segmenting and extracting one or more volumes of at least one
structure in a two-dimensional and a three-dimensional format. The
process of the volumetric extraction renders the one or more
volumes of the at least one structure in a three-dimensional format
that enables a user to study, investigate and analyze the volumes
of at least one structure.
As used herein "volumetric analysis" refers to a process of
analyzing, researching, investigating, and studying the one or more
volumes, shape, orientation, location, and boundaries of the at
least one structure. The volumetric analysis also refers to
analyzing the one or more volumes of the at least one structure and
identifying the cause of the prognosis, diagnosis, atrophy, and
progression and regression of the prognosis.
In an embodiment, the term "volumetric analysis" refers to dealing
with cross-sectional data and seeking measurement of part of or the
total volume of a structure or region of interest. In another
embodiment, the term "volumetric analysis" is any method of
quantitative chemical analysis in which the amount of a substance
is determined by measuring the volume that it occupies or, in
broader usage, the volume of a second substance that combines with
the first in known proportions.
As used herein "family history" refers to the history of medical
events, medical condition, food habitat, brain dominance
information, stress information, micro-ethnicity information, psych
analysis information, symptoms, immunity level, treatments
undergone, medication information, diseases, etc. and the like that
are acquired from family members (e.g., blood relation) of a
patient.
In an embodiment, the term "family history" refers to family
structure and relationships within family, including information
about diseases in family members. Family history provides a ready
view of problems or illnesses within the family and facilitates
analysis of inheritance or familial patterns. In another
embodiment, the term "family history" refers to past occurrences
(of a medical or mental health condition) in family members or past
incidences (of a type of behavior) by family members. Further
"family history" also refers to a record of one's ancestors.
As used herein "patient history" refers to the history of medical
events such as treatments, surgeries, medication that the patient
is undergoing/has undergone till date. The patient history also
refers to medical conditions, food habitat, brain dominance
information, stress information, micro-ethnicity information, psych
analysis information, symptoms, immunity level, treatments
undergone, medication information, diseases, etc. and the like that
are acquired from the patient.
In an embodiment, the term "patient history" refers to case history
of a patient, especially treating the history with correlated
results. Patient history can provide valuable information for the
response, resistance and operative risk of the patient. In another
embodiment, the term "patient history" refers to having relevant
information bearing on their health past, present, and future. The
patient history also comprises medical history, being an account of
all medical events and problems a patient has experienced is an
important tool in the management of the patient.
As used herein "event related potentials (ERP)" refers to one or
more physiological signals that are acquired in response to an
event such as applying at least one stimulus to a patient. The
stimulus may be a tangible stimulus, and a non-tangible
stimulus.
In an embodiment, the term "event related potentials (ERP)" is the
measured brain response that is the direct result of a specific
sensory, cognitive, or motor event. Further, ERP is any stereotyped
electrophysiological response to a stimulus, and includes
event-related spectral changes, event-related network dynamics, and
the like. The stimulus can be a visual stimulus, palpable stimulus,
etc.
In an embodiment, the system comprises a computing unit and a
server communicatively coupled to the computing unit via a
communication network.
In another embodiment, a server may be located in one of a client's
site and a remote place.
In yet another embodiment, the system comprises a dongle associated
with a computing unit to perform at least one of image
segmentation, volumetric extraction, volumetric analysis,
determining atrophy and performing predictive prognosis and
diagnosis.
FIG. 1 illustrates a schematic view of a system, according to one
or more embodiments. The system described herein comprises a
computing unit 102, and a server 104. The computing unit 102 is
communicatively coupled to the server 104 via a communication
network 106. The communication network 106 may be a wired
communication network or a wireless communication network. In an
embodiment, the computing unit 102 is located at a site and the
server 104 is located at a remote place. In another embodiment, the
server 104 and the computing unit 102 is located at the site. The
site may be a hospital, a diagnostic center, a pharmacy, a health
care unit, etc. In an embodiment, the server 104 comprises a black
box. The black box is located locally at the site itself for
securely processing an input and rendering an output (e.g.,
volumes, quantitative volumes, structure-based analysis reports) in
the site itself. In an embodiment, the black box is located locally
at the site to minimize or restrict data anonymization. The server
104 may receive inputs in any combination.
The system may comprise a plug-in device. The plug-in device may
comprise a dongle. The dongle may be associated with the computing
unit 102. In an embodiment, the dongle is communicatively coupled
with the computing unit 102 to perform at least one of volumetric
extraction, volumetric measurements, volumetric analysis,
predicting prognosis, diagnosis, atrophy determination, and
generating a structure-based analysis report. In another
embodiment, the dongle is communicatively coupled with the
computing unit 102 to securely communicate with the server 104 and
perform at least one of the volumetric extraction, the volumetric
measurements, the volumetric analysis, predicting prognosis,
diagnosis, and generating the structure-based analysis report. The
dongle is a key to enable the computing unit 102 to perform at
least one of volumetric extraction, volumetric measurements,
volumetric analysis, atrophy determination, predicting prognosis,
diagnosis, and generating a structure-based analysis report.
The server 104 receives inputs as at least one of an image input, a
text input, and a signal input. The server 104 receives and
analyzes the inputs in any combination (i.e., multivariate pattern
analysis). The server 104 is also capable of receiving different
inputs and performing a multimodal analysis. The image input
comprises one or more first images of a region of interest of an
anatomy. The anatomy may belong to an organism. The organism
comprises one of a human being, an animal, a bird, a mammal and the
like. The one or more first images may comprise one of (a) one or
more computed tomography (CT) scan images, and (b) one or more
magnetic resonance imaging (MRI) scan images, (c) positron emitted
tomography (PET) scan images. The text input comprises demographic
information. The demographic information comprises at least one of
an age, a gender, a race, a micro-ethnicity information and the
like. The text input further comprises a symptom, a medical
condition, etc.
The server 104 may comprise a natural language processing (NLP)
module. The NLP module is configured to check and capture current
clinical symptoms from the text stored on a hospital information
system (HIS). The NLP module may also identify pre indicators like
vitamin deficiency, family history, genetic history, trauma, etc.
The NLP may also extract cognitive analysis from relevant test
analysis like Computerized Cognitive Testing in Epilepsy (CCTE),
Montreal Score, Cambridge Neuro-psychological Test Automated
Battery (CANTAB), Mini Mental State Examination (MMSE), Mini-Cog,
etc.
The signal input comprises one or more physiological signals of a
patient. The one or more physiological signals comprise an
electrocardiography (ECG) signal, an electroencephalogram (EEG)
signal, an Electromyography (EMG), a galvanic skin response (GSR),
a blood pressure, and a heart rate, etc. In an embodiment, the
server 104 can integrate with existing EEG hardware. In another
embodiment, server 104 provides an independent and cloud EEG
service and is available on one-click. The server 104 is also
capable of monitoring the signal input for a predefined period of
time as set by the user and detect for an anomaly (e.g., abnormal
spike, pre-ictal issue, etc.). The signal input may be a prestored
signal or a live signal that is acquired real-time.
FIG. 2 illustrates an exploded view of a server 204, according to
one or more embodiments. The server 204 comprises a memory 206, and
a processor 208. The server 204 also comprises a database 236, and
a networking module 238. The database 236 records and stores a data
repository of segmented images, volumes of the segmented images,
quantitative volumes, categorized with respect to demographic
information (e.g., age, gender, micro-ethnicity information, etc.).
The networking module 238 is configured to communicate with a
computing unit and other hardware or components that the server 204
interacts with. The processor 208 comprises a graphical processing
unit (GPU). The graphical processing unit (GPU) is configured to
support in segmenting images, volumetric analysis, volumetric
extraction, and rendering the one or more volumes in a
three-dimensional format, and at least one anatomical plane. The
server 204 further comprises an input obtaining module 210, a
quality analysis module 212, a user identification data assigning
module 214, a data anonymization module 216, an information linking
module 218, a segmentation module 220, a volume extraction module
222, a volume rendering module 224, a quantitative volume
estimation module 226, a predictive prognosis module 228, a report
compiling module 230, a training module 232, and a retraining
module 234. The processor 208, in association with the
above-mentioned modules, is configured to perform at least one of
image segmentation, volumetric extraction, volumetric measurements,
volumetric analysis, predicting prognosis, diagnosis, atrophy
determination, and generating a structure-based analysis
report.
The input obtaining module 210 obtains inputs as at least one of an
image input, a text input, and a signal input. In an embodiment,
the input obtaining module 210 obtains the inputs as the image
input, the signal input, and the text input (e.g., micro-ethnicity
information). The input obtaining module 210 obtains the inputs in
any combination. The image input comprises one or more first images
of a region of interest of an anatomy. The anatomy may belong to an
organism. The organism comprises one of a human being, an animal, a
bird, and a mammal. The one or more first images may comprise one
of computed tomography (CT), positron emission tomography (PET),
structural magnetic resonance imaging (sMRI), functional magnetic
resonance imaging (fMRI), Diffusion-weighted imaging (DWI),
Diffusion Tensor Imaging (DTI), and magnetic resonance imaging
(MRI) and the like. The text input comprises demographic
information. The demographic information comprises details that
describe characteristics of a patient. The demographic information
comprises at least one of an age, a gender, a race, and a
micro-ethnicity information. The text input comprises at least one
of an age, a race, a gender, a medical condition, a symptom,
clinical history, a patient history, a medical test, medication
information, a physician detail, and a cognitive analysis report,
and the like. The signal input comprises one or more physiological
signals of a patient. The one or more physiological signals
comprise an electrocardiography (ECG) signal, an
electroencephalogram (EEG) signal, an Electromyography (EMG), a
galvanic skin response (GSR), an event related potential (ERP), a
blood pressure, and a pulse rate, etc. In an embodiment, the ERP is
acquired by the server 204 to derive clinical endpoints in at least
one of but not limited to mild cognitive impairment (MCI),
Dementia, Alzheimer, Neurodegeneration, depression, migraines,
stress, concussion, and the like.
The input obtaining module 210 obtains the inputs from at least one
of a computing unit, a Magnetic Resonance Imaging (MRI) scanning
machine, a positron emission tomography (PET) scan machine, a
Computed Tomography (CT) scanning machine and the like. In an
embodiment, the input obtaining module 210 obtains the inputs from
the Magnetic Resonance Imaging (MRI) scanning machine, and the
Computed Tomography (CT) scanning machine directly while scanning
the region of interest of the anatomy (i.e., the server is
integrated with existing scanning machine). In another embodiment,
the input obtaining module 210 obtains the inputs that are obtained
and prestored in the computing unit 202.
The quality analysis module 212 analyzes and determines whether the
quality of the inputs (e.g., the image input, the signal input, the
text input etc.) meets a predefined quality at least one of
qualitatively, and quantitatively. In case of the image input, the
quality analysis module 212 analyzes and determines, in a
quantitative manner, whether a magnetic strength value of the one
or more first images is equivalent to a predefined magnetic
strength value. The predefined magnetic strength value comprises
greater than 1.5 Tesla. In an embodiment, the quality analysis
module 212 performs bias correction on the one or more first images
by compensating bias present in the one or more first images due to
variation in magnetic field and gradient strength of different
(MRI) scanning machines that are currently available in the market.
In another embodiment, the quality analysis module 212 performs
intensity normalization by normalizing the difference in signal
intensities of the one or more first images. The one or more first
images comprise difference in signal intensities due to variation
in acquisition protocol at different sites. In an embodiment, the
quality analysis module 212 enables a user (e.g., radiologist,
technician, etc.) to determine, in a qualitative manner, whether
the one or more first images have the predefined quality (e.g.,
without blur or distortion of pixels, etc.) and can be utilized for
further processing.
In case of the signal input, the quality analysis module 212
analyzes and determines, in a qualitative manner, whether an
amplitude of the one or more physiological signals is equivalent to
a predefined amplitude. In an embodiment, the quality analysis
module 212 performs amplitude normalization by normalizing the
difference in amplitude of the one or more physiological signals.
The one or more physiological signals comprises differences in
amplitudes due to variation in acquisition protocol at different
sites. In an embodiment, the quality analysis module 212 enables
the user to determine, in a qualitative manner, whether the one or
more physiological signals have the predefined amplitude that can
be utilized for further processing.
In case of the signal input, the signal inputs are pre-processed
(such as running through a set of protocol to make it suitable to
the server 204) to filter noises (e.g., bad channel removal) and
artifacts by passing through at least one of a notch filter, a high
pass filter, and a bandpass filter. The signal inputs are then
pre-processed to perform event marking and channel locating within
the inputs. The pre-processing of the signal inputs may include
re-reference and resample of the inputs. The pre-processing of the
signal inputs also includes independent component analysis
(component rejection, if required). The post-processing of the
signal inputs comprises characteristic extraction and
characteristic selection to identify statistically significant
characteristics from the signal inputs using techniques such as
Multivariate time series, Time-frequency (Wavelet transform),
Frequency Domain (Fourier), Time Domain (Principal component
analysis), Independent component analysis (ICA), etc. The
post-processing of the signal inputs further comprises optimal
parameter and characteristic set identification
(e.g.--characteristics shuffle analysis, ranking characteristics,
etc.). The post-processing of the signal inputs further comprises
Classification/Statistical Manipulation (ML) (e.g.--Linear
discriminant analysis (LDA), Multi-layer perceptron (MLP), Support
vector machine (SVM), etc.). The post-processing of the signal
inputs further comprises generating electroencephalogram (EEG)
patterns.
The user identification data assigning module 214 assigns a user
identification data (UID) upon receipt of the inputs. In an
embodiment, the user identification data assigning module 214
assigns a first user identification data (UID) and a second user
identification data (UID) upon receipt of the inputs at a first
site and a second site, respectively. In an embodiment, the user
identification data assigning module 214 assigns a third user
identification data (UID) upon receipt of the inputs at the first
site at a second time. The third user identification data may be
derived from the first user identification data since the inputs
are received at the first site at two different instances.
The data anonymization module 216 anonymizes the inputs received by
discarding the metadata associated with the inputs. The data
anonymization module 216 anonymizes the inputs to remove patient
details from the inputs. The patient details may contribute to
determining the identity or recognizing the patient. Without the
patient details, it would be impossible to detect from whom (e.g.,
which patient, which user, etc.) the inputs are received. In an
embodiment, the data anonymization module 216 anonymizes the inputs
by removing facial detection information and biometrics information
from the inputs. For instance, when the one or more first images of
the same patient received at different instances are combined, it
may contribute to detect/recognize facial information or personal
information of the user. In such a case, the data anonymization
module 216 discards one or more first portions of the one or more
first images. The one or more first portions are the portions that
may assist in recognizing or identifying the user (e.g.,
patient).
The information linking module 218 links information associated
with the first user identification data and the second user
identification data, upon receiving a linking request from the
user. The user, through a user interface, may generate a linking
request to the server 204. For instance, the inputs of the user may
be received by the server 204 at different instances and at
different sites. In such a case, the information linking module 218
links and consolidates the information, associated with the first
user identification data and the second user identification data,
that are obtained at the different sites and the different
instances to give an overall history of the patient to a medical
practitioner.
The segmentation module 220 segments at least one second image of a
structure that resides within the one or more first images. In an
embodiment, the segmentation module 220 segments the at least one
second image using an artificial neural network. The segmentation
module 220 segments the at least one second image of the structure
through one of automatically, semi-automatically, and manually. The
segmentation module 220 segments the at least one second image from
the one or more first images based on structure of the at least one
object (e.g., organ) within the at least one second image. The
segmentation module 220 segments the at least one second image from
the one or more first images through an atlas independent method.
In an embodiment, the segmentation module 220 segments the at least
one second image based on the structure and not by comparing or
aligning with reference image i.e., atlas independent. In an
embodiment, the segmentation module 220 segments the at least one
second image based on the pixels in the structure. In another
embodiment, the segmentation module 220 segments the at least one
second image based on at least one of density/intensity within the
structure. The volume extraction module 222 extracts one or more
volumes of at least one structure from the one or more first
images. The volume extraction module 222 extracts the one or more
volumes by extracting one or more boundaries of the at least one
structure from the one or more first images, and populating one or
more voxels within the one or more boundaries of the at least one
structure using one or more identifiers.
The volume extraction module 222 analyzes the one or more volumes
of the at least one structure and allows the user to determine the
quality of the one or more volumes extracted. The volume extraction
module 222 communicates a signal to the segmentation module 220
when the quality of the one or more volumes extracted is not up to
expected. The segmentation module 220 provides a user interface, in
response to the signal, that allows the user to manually edit and
correct at least one of boundaries, shape, and the one or more
volumes of the structure. The segmentation module 220 then creates
a mask for the structure and allows to populate one or more
identifiers within the structure to correct the one or more volumes
manually. In an embodiment, the mask is created based on the
training provided to the server 204.
The volume rendering module 224 renders the one or more volumes of
the structure in at least one of a two-dimensional format, and the
three-dimensional format. The segmentation module 220 further
renders the one or more volumes of the structure in the at least
one anatomical plane. The anatomical plane comprises at least one
of a horizontal plane, a coronal plane, a sagittal plane, a
parasagittal plane, a transverse plane, an anterior axillary line,
a mid-axillary line, a midclavicular line, a posterior axillary
line and the like. The quantitative volume estimation module 226
estimates one or more quantitative volumes of the at least one
structure based on the pixels/voxels within the structure. The
quantitative volume estimation module 226 provides a numerical
representation of the one or more volumes of the at least one
structure that supports and aids the physician's clinical
impression with quantitative numbers. The numerical representation
i.e., quantitative numbers make the physicians convenient in making
their decisions when compared to the graphical representation or
visual representation of the at least one structure. The
quantitative numbers readily enable the physicians to assess the at
least on structure and predict a prognosis and perform a
diagnosis.
The server 204 creates the data repository (e.g., neuroimaging,
orthopaedic, etc.) using age and gender other than micro-ethnicity
information to create an array of the volumes of different
structures. The server 204 records the volumes of each structure
that comes across and creates the data repository for normal and
abnormal brains for detecting seizures and dementia, MS,
schizophrenia and other anomalies. The server 204 can remove the
effect of age on the volume using a Linear Regression model and
calculating coefficient (e.g., skewness coefficient). The
predictive prognosis module 228 calculates the standard deviation
of the volume of the structures across the cohort. The predictive
prognosis module 228 calculates 25th and 95th percentile of the
standard deviation. The predictive prognosis module 228 calculates
the 25th and the 95th percentile by matching age, gender,
micro-ethnicity information, a medical condition (e.g., epilepsy),
and intracranial volume (ICV) of a patient in the population of
individuals and then deriving the 25th and the 95th percentile
based on matching the age, gender, micro-ethnicity information,
medical condition (e.g., epilepsy), and ICV. The 25th and the 95th
percentile are personalized percentile references in detecting the
predictive prognosis.
The predictive prognosis module 228 determines a feature associated
with the at least one structure based on the one or more volumes
and one or more inputs. The feature comprises at least one of the
one or more volumes of the region of interest (ROI), a cortical
thickness, an atrophy percentage, an asymmetry index score, a
subfield volumetry of the region of interest, annular volume
changes, a progressive supranuclear palsy (psp) index score, a
magnetic resonance perfusion imaging (MRPI) score, a frontal horn
width to intercaudate distance ratio (FH/CC), a medial temporal
lobe atrophy (MTA) score, a global cortical atrophy (GCA) scale,
identification of Intracranial bleeds, hemorrhage, microbleeds and
their volume analysis, a fracture detection, a midline shift
identification, a measurement of the midline shift identification
and the at least one structure with respect to the midline shift
identification, identifying a pathology associated with the at
least one structure, classifying the pathology identified, a tissue
density identification, an infarct identification, a
Penumbra-core-viable tissue identification, classification and
volume calculation, diffusion-weighted imaging (DWI) maps and
apparent diffusion coefficient (ADC) maps of the at least one
structure, perfusion maps comprising resting state functional
magnetic resonance imaging (rsfMRI), an alberta stroke programme
early CT score (ASPECTS) calculation, a collateral detection, a
mismatch ratio calculation, an angiography labeling and/or
annotation, a large vessel occlusion (LVO) detection, an
Hypoperfusion index calculation, Diffusion tensor imaging (DTI)
fiber tracks, neural pathway connectivity maps, correlation between
a signal input, an image input and the text input, classifying the
signal input, identifying a normal signal, identifying an abnormal
signal, identifying a pre-ictal signal, identifying an ictal
signal, extracting symptoms, and grading of condition specific
effects.
The predictive prognosis module 228 also enables to perform
condition specific analysis. The condition specific analysis is
performed by matching a patient's medical condition (e.g.,
epilepsy) with epilepsy population among the population of
individuals and then deriving the 25th and the 95th percentile to
perform proficient predictive prognosis, accurate diagnosis and
comprehensive management. The predictive prognosis module 228
predicts prognosis by analyzing the one or more quantitative
volumes and comparing the one or more quantitative volumes with one
or more reference quantitative volumes (i.e., 25th and 95th
percentile) predominantly based on micro-ethnicity information. The
predictive prognosis module 228 determines and concludes that the
one or more volumes of the patient is normal, when the one or more
volumes of the patient falls between the 25th and the 95th
percentile. The predictive prognosis module 228 also predicts
prognosis and diagnosis by comparing the one or more quantitative
volumes with one or more reference quantitative volumes based on at
least one of an intracranial volume, an age, a gender, a symptom,
and a medical condition. The predictive prognosis module 228 also
predicts prognosis and diagnosis by finding biomarker findings
within the one or more volumes and relating to prestored clinical
observations. Based on the comparison, the predictive prognosis
module 228 predicts the prognosis. The predictive prognosis module
228 is also configured to determine the progression and the
regression of the prognosis over time.
The predictive prognosis module 228 also performs predictive
prognosis by performing volumetric derived analysis such as at
least one of structural analysis, physiological analysis,
functional analysis, and cognitive analysis. For instance, the
structural analysis is performed in case of magnetic resonance
imaging (MRI) inputs. For another instance, the physiological
analysis is performed in case of PET and electroencephalogram (EEG)
inputs. For yet another instance, the functional analysis is
performed in case of Magnetoencephalography (MEG) and fMRI inputs.
For yet another instance, cognitive analysis (i.e., cognitive
performance, cognitive effects, cognitive deficits) is performed in
case of physiological assessment (PA). In view of the above, the
predictive prognosis module 228 may socialize preventive brain
health. The predictive prognosis module 228 identifies biomarkers
from the one or more volumes and relates biomarkers to clinical
presentations and tracks disease progression longitudinally. In one
embodiment, the predictive prognosis module 228 performs structural
volumetric analysis based on 3D MRI correlated with normative
population (specific to micro-ethnicity) when the inputs received
are sMRI. In another embodiment, the predictive prognosis module
228 performs functional mapping of the brain based on an imaging
technique to map different connectivity maps that helps in
understanding disease affected areas and related
cognitive/functional deficits, when the inputs received are fMRI.
In another embodiment, the predictive prognosis module 228 performs
structural as well as perfusion-based analysis of the CT images to
derive a first look into the disease pattern when the inputs
received are CT inputs. In yet another embodiment, the predictive
prognosis module 228 performs white matter tract analysis when the
inputs received are Diffusion tensor imaging (DTI) inputs. The
predictive prognosis module 228 performs predictive prognosis,
diagnosis, and atrophy determination through an atlas independent
method.
The report compiling module 230 generates and compiles a
structure-based analysis report. The structure-based analysis
report comprises at least one of the feature, the one or more
quantitative volumes of the structure, one or more volumes of the
structure, a snippet, volumetric derived analysis, a graphical
representation of prognosis, and the segmented image of the
structure in the at least one anatomical plane. The at least one
feature is rendered in at least one of a two-dimensional (2D)
format, and a three-dimensional (3D) format. The snippet comprises
a brief written description about the medical condition of the
patient. The report compiling module 230 generates and compiles a
first structure-based analysis report and a second structure-based
analysis report for the inputs obtained at a first instance and a
second instance, respectively. The predictive prognosis module 228
predicts the prognosis based on comparison of the first
structure-based analysis report and the second structure-based
analysis report, and the inputs that are obtained at a third
instance. The predictive prognosis module 228 further estimates one
of a progression and a regression of the prognosis associated with
the structure between the first instance and the second instance.
The report compiling module 230 generates and compiles a third
structure-based analysis report based on one of the progression,
and the regression estimated.
In an embodiment, the report compiling module 230 calculates
volumetric derived analysis by using one or more equations. Few of
the sample equations are provided below: ICV=(Volume of
structure/ICV)*100 1)
.times..times..times..times..times..times..times..times..times..times..ti-
mes..times..times..times. ##EQU00001## where LTHC is left
Hippocampus and RTHC is right Hippocampus
.times..times..times..times..times..times..times..times..times..times..ti-
mes..times..times..times..times..times. ##EQU00002## Volume
loss=reference lower limit-volume of structure 4) Total Hippocampus
percentage=((LTHC-RTHC)/ICV)*100 5)
.times..times..times..times..times..times..times..times..times..times..ti-
mes..times..times..times..times..times..times..times..times..times..times.-
.times..times..times..times..times..times..times..times..times..times..tim-
es..times..times..times..times..times..times..times.
##EQU00003##
The training module 232 trains artificial intelligence based neural
network using at least one of the inputs (i.e., the image input,
the text input, and the signal input), the one or more volumes, the
one or more quantitative volumes, the one or more reference volumes
and the one or more reference segmented images. The training module
232 enables the artificial intelligence based neural network to
segment the one or more second images of the at least one structure
from the one or more first images. The training module 232 further
enables the artificial intelligence based neural network to perform
at least one of volumetric extraction, volumetric analysis, atrophy
determination, performing predictive prognosis and diagnosis. In an
embodiment, the training module 232 creates a log using one or more
inputs received from the user while performing manual segmentation
on the one or more first images.
The retraining module 234 retrains the artificial intelligence
based neural network using the log created. The retraining module
234 enables the artificial intelligence based neural network to
automatically segment the one or more second images of the at least
one structure from next time optimized based on the retraining
provided. The retraining module 234 is configured to learn and
perform at least one of the image segmentation, volumetric
extraction, volumetric analysis, atrophy determination, performing
predictive prognosis and accurate diagnosis automatically for any
type of patient belonging to any micro-ethnicity having any type of
medical condition/symptoms in future without any manual
intervention. The predictive prognosis and accurate diagnosis
enable the server to perform a comprehensive management of the
patient's health. The comprehensive management of the patient's
health is performed by performing a predictive prognosis over
time.
For instance, consider the server has predicted a first prognosis
for a condition specific analysis for a first point of time. The
first prognosis is predicted for the first point of time
considering the medication information (e.g., medication that the
patient has intake during the first point of time) of the patient
and other relevant information. The server has also predicted a
second prognosis for a condition specific analysis for a second
point of time. The second prognosis is predicted for the second
point of time considering the medication information (e.g.,
medication that the patient has intake during the second point of
time) of the patient and other relevant information. The server is
also capable of determining deterioration or improvement in at
least one volumetric changes and quantitative volumes by comparing
the first prognosis and the second prognosis. The server determines
the deterioration or the improvement, in terms of percentage,
between the first prognosis and the second prognosis. The server is
then trained with different values of the deterioration or the
improvement for different points of time. The server is then
capable of determining the deterioration or improvement in the
volumetric changes and quantitative volumes for a third point of
time (in future) based on the training provided. The server
determines the deterioration or the improvement for the third point
of time in quantitative values. The quantitative values of the
deterioration or the improvement in the future enables and assists
the physicians to treat/change the medication regime for the
patient accordingly.
The training module 232 and the retraining module 234 enables the
artificial intelligence based neural network to learn and evolve
based on the training and retraining provided. In an embodiment,
the training module 232 and the retraining module 234 enables
creating a data repository for an Indian ethnicity. In another
embodiment, the server 204 records the data repository for a
micro-ethnicity (e.g., a sub region, a sub zone, a city, a state,
etc.). In another embodiment, the server 204 records the data
repository for a macro-ethnicity (e.g., a country, a continent
etc.). As the average volumes for Indian ethnicity were 1122.48 ml
(whole brain) and 1339.75 ml (ICV) as compared to 1222.68 ml (whole
brain) and 1482.87 ml (ICV) in Caucasian's ethnicity; and the age
and gender matched comparison of Indian (group 1) and Caucasian
(group 2) brain and intracranial volumes (ICV) showed significant
difference. The process/method of creating the data repository of
volumes for the Indian ethnicity and its micro-ethnicity seems
significant to implement Artificial intelligence, predict
prognosis, atrophy determination, volumetric extraction, volumetric
analysis, diagnosis and treat patients of the Indian ethnicity.
FIG. 3 illustrates an overview of a system, according to one or
more embodiments. The system obtains inputs as at least one of an
image input, a text input, and a signal input at one or more
instances, and at one or more sites. The image input, the text
input, and the signal input has been described above. The signal
input further comprises clinical data biometrics, and psychological
evaluation. The text input comprises patient data such as
demographic information. Once the inputs are obtained, the system
standardizes the inputs as per the current industry standards by
performing bias correction and normalization.
The system then performs data processing, curation and image
processing to perform segmentation. The segmentation has been
described above. The segmentation can be performed through
automatically, semi-automatically, and manually. The system then
extracts one or more volumes of at least one structure that resides
within one or more first images. The system calculates a feature
associated with the at least one structure. The feature comprises
at least one of the one or more volumes of the region of interest
(ROI), a cortical thickness, an atrophy percentage, an asymmetry
index score, a subfield volumetry of the region of interest,
annular volume changes, a progressive supranuclear palsy (psp)
index score, a magnetic resonance perfusion imaging (MRPI) score, a
frontal horn width to intercaudate distance ratio (FH/CC), a medial
temporal lobe atrophy (MTA) score, a global cortical atrophy (GCA)
scale, identification of Intracranial bleeds, hemorrhage,
microbleeds and their volume analysis, a fracture detection, a
midline shift identification, a measurement of the midline shift
identification and the at least one structure with respect to the
midline shift identification, identifying a pathology associated
with the at least one structure, classifying the pathology
identified, a tissue density identification, an infarct
identification, a Penumbra-core-viable tissue identification,
classification and volume calculation, diffusion-weighted imaging
(DWI) maps and apparent diffusion coefficient (ADC) maps of the at
least one structure, perfusion maps comprising resting state
functional magnetic resonance imaging (rsfMRI), an alberta stroke
programme early CT score (ASPECTS) calculation, a collateral
detection, a mismatch ratio calculation, an angiography labeling
and/or annotation, a large vessel occlusion (LVO) detection, an
Hypoperfusion index calculation, Diffusion tensor imaging (DTI)
fiber tracks, neural pathway connectivity maps, correlation between
a signal input, an image input and the text input, classifying the
signal input, identifying a normal signal, identifying an abnormal
signal, identifying a pre-ictal signal, identifying an ictal
signal, extracting symptoms, and grading of condition specific
effects.
The system then quantifies the one or more volumes and provides a
numerical representation of the one or more volumes. The system
then normalizes the one or more quantitative volumes so that the
one or more quantitative volumes can be used for prognosis,
diagnosis, atrophy determination, and treatment purposes. Based on
the quantitative numbers, the physician can determine the
percentage of atrophy over time.
The one or more volumes, the one or more quantitative volumes, etc.
are recorded as a data repository. The data repository is provided
to a wide range of healthcare practitioners to aid in their
assessment of a patient's prognosis. The one or more volumes and
the one or more quantitative volumes also assists the wide range of
healthcare practitioners for performing at least one of an
objective analysis, and a subjective analysis. The objective
analysis helps in analyzing and studying characteristics of the
least one object (i.e., structure). The subjective analysis helps
in analyzing and studying characteristics of the least one subject
(e.g., patient). The system also provides services such as imaging
biomarkers and predictive analytics (i.e., predictive prognosis,
diagnosis and atrophy). The system also generates a structure-based
analysis report comprising at least one of the feature, the one or
more volumes of the at least one structure represented in a
three-dimensional format and an anatomical plane which aids to
perform subjective and/or objective analysis. The system also
creates a localized data repository comprising one or more volumes
and one or more quantitative volumes of at least one organ
categorized with respect to micro-ethnicity, age, gender, race,
ICV, and the like. The appropriate users may be structure-based
analysis report consumers, software and enhanced service consumers,
and data insights consumers. The system comprises a layer having at
least one of neural networks, deep learning and artificial
intelligence algorithms and data analytics. The system further
comprises a layer of data integrity, cyber security and compliance.
The system further comprises a layer of cloud-based user interface
platform for anonymization and clinical and medical imaging data
management.
FIG. 4 illustrates a multivariate pattern analysis performed by a
system, according to one or more embodiments. The server receives
inputs as at least one of an image input 440, a text input 444, and
a signal input 442. The image input 440 comprises one or more first
images of a region of interest of an anatomy. The anatomy may
belong to an organism. The organism comprises one of a human being,
an animal, a bird, a mammal, and the like. The one or more first
images may comprise one of (a) one or more computed tomography (CT)
scan images, (b) one or more magnetic resonance imaging (MRI) scan
images, (c) one or more positron emitted tomography scan images
(PET), (d) one or more functional magnetic resonance imaging (fMRI)
scan images, (e) one or more structural magnetic resonance imaging
(fMRI) scan images, (f) Diffusion Tensor Imaging (DTI), and (g)
Diffusion-weighted imaging (DWI). The text input 444 predominantly
comprises demographic micro-ethnicity information. The text input
444 further comprises information of at least one of an age, a
gender, a race, an intracranial volume (ICV), a symptom, a medical
condition, clinical history, psych analysis information, stress
information, brain dominance information, food habitat information,
family history, clinical history, etc. The signal input 442
comprises one or more physiological signals such as at least one of
but not limited to electrocardiography (ECG) signal, an
electroencephalogram (EEG) signal, an Electromyography (EMG), a
galvanic skin response (GSR), a blood pressure, and a heart rate,
etc. The signal input 442 may be one or more physiological signals
that are recorded and/or pre-stored. In an embodiment, the signal
input may be one or more physiological signals that are acquired in
real-time.
The server receives the inputs in any combination to perform
multivariate pattern analysis 446. The server receives the inputs
in any combination such as (i) combination comprising the image
input and the text input, (ii) combination comprising the signal
input and the text input, (iii), combination comprising the image
input, the signal input, and the text input. Since the
micro-ethnicity information serves as a major distinguishing factor
between groups of peoples to perform structure-based analysis,
volumetric extraction, volumetric analysis, atrophy determination,
quantification, and perform predictive prognosis and diagnosis, the
text input comprises predominantly the micro-ethnicity information.
The text input further comprises intracranial volume (ICV), age,
gender, race, medical symptoms, and the like. The text input such
as the micro-ethnicity information, age, gender, ICV and the like
may have an impact in the volumes of the at least one structure in
the one or more first images.
FIG. 5 illustrates a method of structure-based analysis report
generation, according to one or more embodiments. The method of
structure-based analysis report generation comprises steps of
capturing and/or obtaining inputs from a site (step 548),
pre-processing of the inputs (step 550), segmentation (step 552),
volume extraction (step 554), quality check (step 556), and
reporting (step 558). The pre-processing step 550 further comprises
discarding metadata associated with the inputs by converting the
inputs from a first format (e.g., Digital Imaging and
Communications in Medicine (DICOM) format) to a second format
(e.g., Neuroimaging Informatics Technology Initiative (NIfTI)
format). The inputs are verified for meeting industry standards and
quality. In case of an image input, one or more first images are
then verified for magnetic strength value having more than 1.5
Tesla associated with the one or more first images. One or more
second images of at least one structure resided within the one or
more first images are then segmented at step 552. One or more
volumes of the at least one structure is extracted and rendered to
a user to perform structure-based analysis (i.e., volumetric
analysis) at step 554. At least one feature is also determined at
step 554 based on the one or more volumes extracted and the one or
more inputs received.
The feature comprises at least one of the one or more volumes of
the region of interest (ROI), a cortical thickness, an atrophy
percentage, an asymmetry index score, a subfield volumetry of the
region of interest, annular volume changes, a progressive
supranuclear palsy (psp) index score, a magnetic resonance
perfusion imaging (MRPI) score, a frontal horn width to
intercaudate distance ratio (FH/CC), a medial temporal lobe atrophy
(MTA) score, a global cortical atrophy (GCA) scale, identification
of Intracranial bleeds, hemorrhage, microbleeds and their volume
analysis, a fracture detection, a midline shift identification, a
measurement of the midline shift identification and the at least
one structure with respect to the midline shift identification,
identifying a pathology associated with the at least one structure,
classifying the pathology identified, a tissue density
identification, an infarct identification, a Penumbra-core-viable
tissue identification, classification and volume calculation,
diffusion-weighted imaging (DWI) maps and apparent diffusion
coefficient (ADC) maps of the at least one structure, perfusion
maps comprising resting state functional magnetic resonance imaging
(rsfMRI), an alberta stroke programme early CT score (ASPECTS)
calculation, a collateral detection, a mismatch ratio calculation,
an angiography labeling and/or annotation, a large vessel occlusion
(LVO) detection, an Hypoperfusion index calculation, Diffusion
tensor imaging (DTI) fiber tracks, neural pathway connectivity
maps, correlation between a signal input, an image input and the
text input, classifying the signal input, identifying a normal
signal, identifying an abnormal signal, identifying a pre-ictal
signal, identifying an ictal signal, extracting symptoms, and
grading of condition specific effects.
The at least one feature is rendered. The one or more volumes are
also rendered in at least one of a three-dimensional format, and a
two-dimensional format. The one or more volumes are also rendered
in at least one anatomical plane. The one or more volumes are then
sent for quality control, at step 556, to provide an optimum
quality and optimum shape of pictorial representation of the one or
more volumes, prior to compiling a structure-based analysis report.
At step 558, the server generates a structure-based analysis report
that gives insights to the physicians, caregivers, radiologists,
researchers, etc. The system also performs integrated analysis by
using at least one of the text input, and the signal input with the
image input acquired from the same patient to provide optimum
accuracy in predictive prognosis, diagnosis, and atrophy
determination. The structure-based analysis report aids physicians,
doctors, and medical practitioners in their assessment of a
patient's prognosis.
FIG. 6 illustrates users of a system, according to one or more
embodiments. The system extracts and renders one or more volumes of
at least one structure in at least one of a three-dimensional (3D)
format, and an anatomical plane. Since the rendered volumes are
utilized for research and study purposes, the one or more volumes
rendered should be rendered in optimized quality. Further the one
or more rendered volumes are utilized for treatment purposes, the
system may be utilized and accessed by the users at a site. The
site may comprise a hospital, a diagnostic center, a health care
unit, etc. The system is also utilized in creating a database with
respect to a micro-ethnicity (e.g., an Indian micro-ethnicity)
comprising one or more features, one or more volumes, one or more
quantitative volumes, age, gender, micro-ethnicity information,
etc. At least for the above use cases, the system may be utilized
by the users such as radiologist, technician, manager, analyst,
doctors, students, researchers, physicians, etc.
The technicians may upload, study, attach documents, view studies,
view reports, etc., to finalize and pass on to the next stage. The
analyst may view assigned studies, add clinical story, view study,
perform manual segmentation, submit to the manager after quality
check, etc. The admin may create/edit sites, and create/edit users.
The manager may view studies, assign studies to an analyst, prepare
reports, send to a radiologist, quality check finalization, and
finalize reports. The radiologist may view studies and segmented
images, view reports, and perform QC review.
FIG. 7 illustrates a process flow of a system, according to one or
more embodiments. The process flow shown here splits the flow under
different sections for illustrative purposes. The process flow is
split under different sections such as onboarding users, operations
from sites, operations, and quality control and delivery. Under the
onboarding users' section, the user is enabled to perform initial
settings (e.g., register, sign-up, etc.) and provide access to a
server. Performing Initial settings comprise providing user
information and logging in. Once the access has been given, the
server depicts demos such as trial cases (3-5). The trial cases
enable the server in understanding variations associated with each
user and knowing what should not be changed at that point. User ID
is then generated for a site at which the user is operating. In an
embodiment, User ID is generated for each user, in addition to the
user ID generated for the site. When there are multiple diagnostic
and hospital chains, the user ID is generated per location as
described above. The server also provides two user IDs for each
location. For example, the server provides a first user IDs for a
technician and a second user ID for a radiologist at the same
location.
Once the user IDs have been generated, the server enables the user
to perform the operations, such as case uploads, from sites. The
server enables the user to login to a web portal using the user ID
generated. The user must upload a consent form. The user can then
scan and upload inputs (e.g., one or more first images, one or more
physiological signals, text inputs). The user is enabled to enter
or provide other information to the server (portal or platform).
Once the requisite and mandatory information has been uploaded, the
uploaded case will reflect online in a cloud (such as Amazon Web
Services (AWS.RTM.)) to all users who have access as per privileges
given. For instance, when the case is uploaded the doctor can
access his patient's case and perform the assessment. The doctor
cannot access the case history of other doctor patients.
Once the case has been uploaded from the sites, the server enables
the user to perform operations such as management operations. Under
the operations, the user such as manager may assign the case to an
analyst. Upon assigning the case to the analyst, the analyst gets
access to it. The analyst then performs quality assurance/quality
control. Once the quality assurance has been complied, an image
segmentation is performed (e.g., automatically, semi-automatically,
or manually) to segment at least one structure that resides within
the one or more first images. Data retrieval (e.g., logs) from
algorithms is then performed to train the neural networks.
An analysis report is then generated and compiled. The analysis
report comprises at least one of the feature, volumetric analysis
such as one or more segmented images in at least one anatomical
plane and the one or more quantitative volumes of the at least one
structure. The analysis report further comprises volumetric derived
analysis such as segmentation prediction, volume calculation,
reference range calculation, atrophy, ICV volume, Volume loss, etc.
The analysis report is then passed on to the quality control.
Under the quality control and delivery, the server enables a second
analyst to check the one or more segmented images and make
appropriate changes, as necessary. The compiled report is then
cross checked by a third analyst to ensure quality and make
appropriate changes, as necessary. The final report is then checked
by a manager. The final report is then sent to the
radiologist/principal investigator. The final report readily
provides quantitative numbers, volumes of at least one organ,
reference quantitative volumes, a snippet regarding prognosis, etc.
to the physicians. The final report aids the physicians in
predictive prognosis and diagnosis.
FIGS. 8 and 9 illustrate a system architecture, according to one or
more embodiments. The system architecture depicts a user
interacting with at least one DICOM server (e.g., ORTHANC) through
a web server. The DICOM server is communicatively coupled to other
database servers. DICOM studies are uploaded directly to the DICOM
servers by DICOM users or through the web server by non-DICOM
users. The high-level design shown in FIG. 8 illustrates that both
the DICOM and non-DICOM users interact with the system. The DICOM
server provides the one or more first images, the one or more
second images, and the segmented structures to a web portal for all
purposes including investigation, study, treatment and other two
dimensional or three-dimensional model creation purposes. The DICOM
servers let the users focus on content of the DICOM files, hiding
the complexity of the DICOM format and of the DICOM protocol. The
DICOM server provides imaging contents to the web server for all
purposes and will be the primary source for the study list.
Once the DICOM studies are populated, the documents are uploaded to
S3 bucket (e.g., AWS.RTM. S3 bucket) based on user ID. The S3
bucket is configured to store physical image files for the
application. Other metadata about clinical history and other
demographic information will be uploaded in a second database
(e.g., MYSQL database) which will be the portal main database and
not accessible outside. The subsequent workflows about study
management will be handled in the second database. In an
embodiment, the system uses a viewer (e.g., papaya viewer) to view
and edit the one or more first images.
FIG. 10 illustrates a workflow, according to one or more
embodiments. The process flow describes a sequential process. A
server allows the user (e.g., admin) to create an account for a
site and an account for the user, at step 1002. Once the account
for the user is created, the user is enabled to upload a case and a
patient history form through the account created, at step 1004. The
user is also enabled to upload inputs such as at least one of an
image input, a text input, and a signal input. At step 1006, upon
successful uploads, the inputs are populated on a worklist of a
responsible user by the server. A manager does quality control by
verifying the quality of the inputs and accepts and rejects the
inputs accordingly. At step 1008, the server enables the manager to
assign the case to an analyst for quality assurance/quality
control. Once the quality assurance is performed, the inputs are
passed on for processing. At step 1010, in case of the image input,
one or more second images of at least one structure that resides
within one or more first images are segmented by an automatic
segmentation application programming interface (API).
The segmented images are then sent to the analyst to quality
control and work on the image input. In case of the signal input,
one or more physiological signals are then sent to the analyst to
quality control and work on the signal input. At step 1012, the
segmented images are sent to the radiologist for feedback. If the
Radiologist suggests changes, the case goes back to the analyst for
processing (e.g., manual segmentation, volumetric extraction,
volumetric analysis, atrophy, and quantification). The process is
repeated until there is no negative feedback from the radiologist.
At step 1014, one or more features associated with the at least one
structure, one or more volumes of the at least one structure, one
or more quantitative volumes, and reference ranges (e.g., 25th
percentile and 95th percentile) for the at least one structure is
then calculated using the volumetric API by the server once the
segmentation is over. At step 1016, the final volumes and the
reference ranges are calculated by the volumetric API and are
populated in a structure-based analysis report by the server. At
step 1018, the generated report will be approved by an admin and it
will be entered in a database. At step 1020, the structure-based
analysis report and a patient history will be made available to all
user accounts by the server as per privileges assigned.
FIG. 11a-11e illustrate a process of segmentation of Hippocampus,
according to one or more embodiments. The process of segmentation
of the Hippocampus comprises the following technical steps to be
executed. A server enables a user to upload one or more first
images of a region of interest (i.e., skull) to an ITK snap layer
of the server. The ITK snap layer of the server allows the user to
navigate three-dimensional medical images, manually delineate
anatomical regions of interest, and perform automatic image
segmentation. The server enables the user to import a label file
for the Hippocampus. Hippocampus label file comprises predefined
RGB values. In an embodiment, the predefined RGB values of the
Hippocampus label file assigned are R-255, G-182, B-139.
Once the Hippocampus label file is imported, the Hippocampus
structure is segmented automatically, using artificial
intelligence, from the one or more first images. Then a contrast
inspector drop-down tab is accessed as shown in FIG. 11a via a user
interface, rendered by the server, to adjust the contrast so that
grey matter (GM) and white matter (WM) differentiation is optimum.
The one or more segmented images are then rendered in at least one
anatomical plane such as a sagittal plane, an axial plane, and a
coronal plane to readily enable a user to visualize the Hippocampus
in the at least one anatomical plane and identify a location,
position and shape of the Hippocampus. The Hippocampus comprises a
right Hippocampus and a left Hippocampus. Upon determining, when
the one or more segmented images do not comprise optimized quality
in terms of shape, boundary and volume of at least one structure,
the segmented images can be further manually edited by performing
manual segmentation.
The server, via the user interface, enables the user to move to an
image slice using a sagittal plane when the right Hippocampus
disappears. The server enables the user to use a "polygon" tool on
a main toolbar as shown in FIG. 11b. The "polygon" tool, upon
selecting, enables the user to perform the manual segmentation by
drawing and filling polygons in orthogonal image slices. In an
embodiment, the manual segmentation can be done individually in the
anatomical plane. The manual segmentation, via the polygon tool,
enables the user to add points to the polygon and edit the
completed polygon. The "polygon" tool enables the user to zoom in
and out (hold and drag) to view any specific portion of the
Hippocampus. The "polygon" tool further enables the user to place
and move 3D cursor, scroll through image slices and scroll through
image components to view, edit and correct the volume, shape and
structure of the Hippocampus. The server further provides an
"active label" tool under "segmentation label". Under the "active
label" drop-down tool, the user is enabled to select an appropriate
label (i.e., right Hippocampus in this instance) as shown in FIG.
11c. The server further enables the user to select the "paint over"
tool as all labels.
The server enables the user to choose opacity so as not to
obscure/hide tissue boundaries. In an embodiment, the opacity
ranges between 15-30. The server enables the user to outline the
outermost border of the Hippocampus using the image slice chosen as
shown in FIG. 11d. In an embodiment, a first color (e.g., pink) is
used for an active polygon and a second color (e.g., red) stands
for completed polygon. The server further enables the user to
retrace borders of the Hippocampus and detect any missing pixels or
voxels by zooming in. The server provides a "brush" tool. The
"brush" tool enables the user to edit and add the missing
pixels/voxels by selecting an appropriate brush (e.g., round brush,
square brush) and appropriate bush size. If the edits have been
done more than the actual voxels/pixels (i.e., in case of over
estimation), the server enables the user to select the "active
label" as "clear label" and edit the voxels/pixels.
In an embodiment, the Hippocampus was defined to comprise
subiculum, Ammon's horn (CA1-CA4), dentate gyrus, and associated
white matter tracts (alveus, fimbria). Ammon's horn within the
posterior uncus was also included. In an embodiment,
disarticulation of the Hippocampal head from the amygdala and
uncinate gyms on the most anterior sections was aided by
recognizing the undulating contour of the pes digitations and by
the fact that the alveus provides a high signal intensity (white
matter) marker defining superior border of the head of the
Hippocampus where it directly abuts the overlying.
The server also enables the user to segment the left Hippocampus
and extract volumes of the left Hippocampus. Once the right
Hippocampus is completed repeat the above steps for the left
Hippocampus. To proceed with the left Hippocampus, the server
enables the user to change the "active label" as left Hippocampus
(Lt HC) before starting left HC. The segmented image and extracted
volumes of both the left Hippocampus (Lt HC) and the right
Hippocampus (Lt HC) are shown in FIG. 11e.
Once both the segmentation and volume extraction of the right HC
and left HC are complete, the server enables the user to save the
one or more first images, workspace, mesh and the one or more
segmented images with a patient id name. The server enables the
user to place the directory location for the image files in the
patient ID folder. The server enables the user to capture
screenshots/save the segmented images at all anatomical planes in
such a way that both the Lt HC and Rt HC are displayed well with
labels. The server further allows the user to capture screenshots
of the segmented image in a three-dimensional image format with
both the left HC and right HC zoomed well.
The server enables the user to check for errors, starting from the
tail of the hippocampus in the axial plane. It is to be noted that
Hippocampus do not include the Gyms. The server enables the user to
check the boundary using the cursor and check in all three
anatomical planes shown. Since the tail of the Hippocampus is close
to Thalamus, check the voxels/pixels in the sagittal plane and the
coronal plane for better visualization for separation between the
Thalamus and the Hippocampal tail. Fimbria is to be included with
the Hippocampus. Fimbria is visualized as a hyperintense layer on
the surface of the Hippocampus. Fimbria is also visualized in the
sagittal plane and the axial plane attaching the head of the
Hippocampus to the Amygdala. Further the head of the Hippocampus
and the Amygdala is distinguished by a slight hyperintense boundary
of the Hippocampal head. The slight hyperintense boundary of the
Hippocampal head is visualized in the anatomical planes. The server
allows the user to trace the slight hyperintense boundary of the
Hippocampal head using the "cursor chase" option and add the slight
hyperintense boundary in the image segmentation.
FIG. 12a-12k illustrate a process of segmentation of Ventricles,
according to one or more embodiments. The process of segmentation
of the Ventricles comprises the following technical steps to be
executed. While performing the segmentation, a server enables a
user to select the "active label" as "Ventricles". In an
embodiment, RGB values assigned for the Ventricles are: R-181,
G-176, and B-22. The server further enables the user to segment the
Ventricles in at least one of Automatic segmentation,
Semi-automatic segmentation, and Manual segmentation. As the
Ventricles are bounded by complex tissue matters, the Ventricles is
mostly segmented using Contour Segmentation i.e., Semi-automatic
segmentation. The contour segmentation allows the user to select a
region of interest from uploaded one or more first images (e.g.,
skull) for semi-automatic active contour segmentation and start the
semi-automatic segmentation as shown in FIG. 12a. The contour
segmentation enables the user to adjust the boundaries of the
region of interest. Once the `active label` is assigned as
`Ventricles`, the server renders the one or more first images in at
least one anatomical plane and a three-dimensional format. The
server enables the user to identify the Ventricles in the at least
one anatomical plane rendered by the server.
The Ventricles can be identified by understanding the structure of
the Ventricles. Cerebral ventricular system is made up of four
ventricles that comprises two lateral ventricles (one in each
cerebral hemisphere), a third ventricle in a diencephalon, and a
fourth ventricle in a hindbrain. The lateral ventricle is a
C-shaped cavity situated within each cerebral hemisphere. The two
lateral ventricles are separated from each other by a thin vertical
sheet of nervous tissue called septum pellucidum covered on either
side by ependyma. The two lateral ventricles communicate with the
third ventricle through the interventricular foramen of Monro. Each
of the lateral ventricles is made up of a central part (body) and
three horns (cornua) namely the anterior horn, posterior horn, and
inferior horn. Anterior wall is formed by the posterior surface of
the genu of corpus callosum and the rostrum. The roof is formed by
an inferior surface or anterior part of the body of the corpus
callosum. Medial wall is formed by the septum pellucidum. The floor
is formed majorly by the head of the caudate nucleus, while a small
portion on the medial side is formed by the upper surface of the
rostrum of the corpus callosum. The roof and lateral wall of the
posterior horn are formed by the sheet of fibers of corpus callosum
known as tapetum. This separates the posteriorly sweeping optic
radiation from the cavity of the posterior horn. The medial wall
has 2 bulges. In the upper part, it is formed by the fibers of the
occipital lobe sweeping backward known as forceps major and is
referred to as the bulb of the posterior horn. The second elevation
below this is called calcar avis and corresponds to the in-folding
of the anterior part of calcarine sulcus.
The inferior horn forms a curve around the posterior end of the
thalamus, descending posterior laterally and then anteriorly into
the temporal lobe. The area where inferior horn and posterior horn
diverge is called collateral trigone or atrium. Laterally, the roof
is covered by the inferior surface of the tapetum of the corpus
callosum and medially by the tail of the caudate nucleus and stria
terminalis. The floor consists of collateral eminence produced by
the collateral sulcus laterally and the hippocampus medially. The
fibers of the hippocampus form a thin layer of white matter called
alveus that covers the ventricular surface and converge medially to
form the fimbria. Most medially on the floor lies the choroid
plexus passing through the choroid fissure.
The third ventricle is a median slit-like cavity situated between
the two thalami and part of the hypothalamus. In the anterosuperior
aspect, the third ventricle communicates with the lateral
ventricles while on its posteroinferior aspect the third ventricle
communicates with the fourth ventricle through the cerebral
aqueduct of Sylvius. The space of the third ventricle is lined by
ependyma and is traversed by a mass of grey matter called
interthalamic adhesion or Massa intermedia, located posterior to
the foramen of Monroe and connects the two thalami. The fourth
ventricle is bounded anteriorly by the pons and cranial half of the
medulla and posteriorly by the cerebellum. The fourth ventricle
appears triangular on the sagittal section and rhomboidal on the
horizontal section. Superiorly, the fourth ventricle is continuous
with the cerebral aqueduct while inferiorly the fourth ventricle is
continuous with the central canal of the spinal cord.
Once the Ventricles are identified, the server enables the user to
mark a region of interest covering the Ventricles and check whether
the Ventricles is covered in the anatomical planes as shown in FIG.
12b. The server provides a "segment 3D" tool that allows the user
to start semi-automatic segmentation. The user upon clicking the
"segment 3D" tool a toolbar appears. The toolbar provides a
"pre-segmentation" tool. The "pre-segmentation" tool provides at
least one of "thresholding", "classification", "clustering", and
"edge attraction" as shown in FIG. 12c. The user can select any of
the four tools under the "pre-segmentation" tool. For example, the
"classification" tool is selected by the user as shown in FIG.
12d.
Before starting the "classification" type segmentation, the server
enables the user to choose a label indicating "Ventricles" under
the "Segmentation labels" tool. The "Segmentation label" tool is
used to record and save information (e.g., volumes, quantitative
volumes, boundaries, manual edits performed to the segmentation,
etc.) obtained as a result of the segmentation performed on the at
least one structure (i.e., the ventricles). The server further
provides a "brush" tool that allows the user to select appropriate
brush (e.g., round brush, square brush) and appropriate brush size
to mark the Ventricles under the "Ventricles" label of the
"Segmentation labels". The server further allows the user to
differentiate tissue samples and the Ventricles. The different
tissue samples comprise white matter (WM) and grey matter (GM). The
server allows the user to select the "Segmentation labels" as
"Brain" and mark the tissue samples such as the WM and GM.
The server further provides a "train classifier" tool that allows
the user to train a classifier by clicking on the "train
classifier" tool once the tissue samples are marked appropriately.
The server further renders a speed image that shows the
classification. The "train classifier" assigns a probability value
to a voxel belonging to the "foreground" class vs. a voxel
belonging to all other classes i.e., the Ventricles and the Brain.
Once the classifier is trained using the manual segmentation (i.e.,
marking and differentiating the tissue samples and the Ventricles)
the classifier automatically segments at least one structure within
the one or more first images of a different patient in future based
on micro-ethnicity information, age, race, gender, patient history,
clinical history, medical condition, symptoms, brain dominance
information, stress information, food habitat information, psych
analysis information, etc. "Foreground class" tab below shows the
labels (e.g., Ventricles and Brain) of structures within the region
of interest for which the boundaries are marked and differentiated.
Upon selecting the "Ventricles" label under the "Foreground class",
the server renders the Ventricles in the anatomical planes and in
the speed image, as shown in FIG. 12e. Similarly, upon selecting
the "Brain" label under the "Foreground class", the server renders
the whole brain in the anatomical planes and in the speed image as
shown in FIG. 12f. The server provides a "Next" tab to complete and
finalize the segmentation process.
FIG. 12g illustrates the segmentation of the Ventricles performed
using a thresholding method. Under the "pre-segmentation" tool
select the "thresholding" tool. The server provides a "More" tab
that provides a speed image generation window (as shown in FIG.
12c). Speed image generation renders a lower threshold slider, an
upper threshold slider, and a smoothness slider that allows the
user to adjust an upper threshold value, a lower threshold value
and smoothness value. The sliders are adjusted to remove the white
matter (WM) from the selection/classification area properly. In an
embodiment, the lower threshold value is adjusted to zero while the
upper threshold value is adjusted to a minimum value to remove the
white matter from the classification/selection area. Once the white
matter is removed, the server provides a "Next" tab to finalize and
submit the threshold levels.
The server then provides an "Add bubble at cursor" tool that allows
the user to populate bubbles of appropriate sizes exactly in the
ventricles in at least three anatomical planes to exactly extract
the volume of the Ventricles as shown in FIG. 12h. The server also
provides a "Bubble radius" slider that allows the user to vary the
size of the bubbles. The server further provides an "active
bubbles" drop down menu that shows the bubbles and its radius that
are active. The server allows the user to add a sufficient number
of bubbles in the Ventricles. The server also provides a "delete
active bubbles" tool that enables the user to delete appropriate
bubbles and populate the bubbles exactly only within the boundaries
of the Ventricles. The server provides a "Next" tab to finalize the
volume extraction.
The server provides a "continuous update" tool that enables to
continuously update contour evolution. The server further provides
a "play" that allows the user to play and pause Active Contour
Evolution as shown in FIG. 12i. The server further provides a
"finish" tab as shown in FIG. 12j that allows the user to submit
when the active contour evolution is done. The server allows the
user to change the "active label" to "clear label" and edit the
voxels when the active contour evolution goes out of the boundaries
of the Ventricles. The server allows the user to edit the voxels by
accessing the "brush" tool and selecting appropriate brush and
appropriate brush size. The server allows the user to change the
"active label" to "Ventricles" and edit the voxels/pixels when the
active contour evolution has not reached any part of the
Ventricles. The server allows the user to edit the voxels/pixels by
accessing the "brush" tool and selecting appropriate brush and
appropriate brush size. The server captures and records actions
performed by the user under the "active label".
Once both the segmentation and volume extraction of the Ventricles
are complete, the Ventricles are rendered in at least one
anatomical plane and a three-dimensional format as shown in FIG.
12k. The server enables the user to save the one or more first
images, the workspace, the mesh and the one or more segmented
images with a patient id name. The server also places directory
location for the image files in the patient ID folder. The server
enables the user to capture screenshots of the segmented images at
all anatomical planes in such a way that Ventricles are displayed
well with labels. The server further allows the user to capture
screenshots of the segmented image in a three-dimensional image
format with the Ventricles zoomed in and displayed well. Boundaries
of the Ventricles comprises the following. The lateral ventricles
temporal horn is separated by the fimbriae. Segmentation to be done
according to separation. The marked area boundaries are the defined
anterior and posterior boundaries for the third ventricles.
FIG. 13a-13h illustrate a process of segmentation of a Whole Brain,
according to one or more embodiments. The process of segmentation
of the Whole Brain comprises the following technical steps to be
executed. While performing the segmentation, a server enables a
user to select the "active label" as "Whole Brain". In an
embodiment, RGB Values assigned for the Brain are: R-197, G-239,
and B-91. The server further enables the user to segment the Whole
Brain in at least one of Automatic segmentation, Semi-automatic
segmentation, and Manual segmentation. As the Brain is bounded by
complex tissue matters, the Brain is mostly segmented using Contour
Segmentation mode i.e., Semi-automatic segmentation. The contour
segmentation allows the user to select a semi-automatic "active
contour segmentation" tool and start the semi-automatic
segmentation as shown in FIG. 13a. The contour segmentation enables
the user to adjust the boundaries of the region of interest
covering the Whole Brain. Once the `active label` is assigned as
"Whole Brain", one or more first images (e.g., skull) are rendered
in at least one anatomical plane and a three-dimensional format.
The server enables the user to identify the Whole Brain in the at
least one anatomical plane.
Once the Whole Brain is identified, the server enables the user to
mark a region of interest covering the Whole Brain and check
whether the Whole Brain is covered in the anatomical planes as
shown in FIG. 13b. The server provides a "segment 3D" tool that
allows the user to start semi-automatic segmentation. The user upon
clicking the "segment 3D" tool a toolbar appears. The toolbar
provides a "pre-segmentation" tool. The "pre-segmentation" tool
provides a "thresholding", "classification", "clustering", and
"edge attraction". The user can select any of the four tools under
the "pre-segmentation" tool. For example, the "classification" tool
is selected by the user as shown in FIG. 13c.
Before starting the "classification" type segmentation, the server
enables the user to choose a label indicating "Whole Brain" under
the "Segmentation labels" tool. The "Segmentation label" tool is
used to record and save information (e.g., volumes, boundaries,
manual edits performed to the segmentation, etc.) obtained as a
result of the segmentation performed on the at least one structure
i.e., Whole brain. The server further provides a "brush" tool that
allows the user to select appropriate brush (e.g., round brush) and
appropriate brush size to mark the Whole Brain under the "Whole
Brain" label of the "Segmentation labels". The Whole Brain
structure marked comprises brain structures (e.g., grey matter
(GM), white matter (WM), Midbrain, Pons, Medulla). The
"classification" tool allows the user to classify between Brain and
Intracranial Volume (ICV) under two labels "Whole Brain" and "ICV".
The "Whole Brain" label is used to classify between white matter
and grey matter. The "ICV" label is used to classify between dura,
skull bone, Ventricles or cerebrospinal fluid (csf). In an
embodiment, if there is an error, the "classification" tool further
allows the user to add a third label as "Ventricles" to classify
the Ventricles separately. The different tissue samples comprise
white matter (WM) and grey matter (GM). The server allows the user
to mark the tissue samples such as the WM and GM.
The server further provides a "train classifier" tool that allows
the user to train the classifier by clicking on the "train
classifier" tool as shown in FIG. 13c. The server further renders a
speed image that shows the classification between the whole brain,
skull and ICV. The "train classifier" assigns a probability value
to a voxel belonging to the "foreground" class vs. belonging to all
other classes. Once the classifier is trained using the manual
segmentation (i.e., marking and differentiating the tissue samples,
Brain, ICV, and the Ventricles) the classifier automatically
segments at least one structure within the one or more first images
of a different patient in future based on micro-ethnicity
information, age and gender. "Foreground class" tool below shows
the labels (e.g., Skull, ICV and Brain) of structures within the
region of interest for which the boundaries are marked and
differentiated. Upon selecting the "Whole Brain" label under the
"Foreground class", the server renders the Whole Brain in the speed
image, as shown in FIG. 13d. Similarly, upon selecting the "Skull"
label under the "Foreground class", the server renders the Skull in
the speed image. The server provides a "Next" tool to complete the
segmentation process.
The server provides an "Add bubble at cursor" tool that allows the
user to populate bubbles of appropriate sizes exactly in the Whole
Brain in at least three anatomical planes to exactly extract the
volume of the Whole Brain as shown in FIG. 13e. Further the server
provides a "Bubble radius" slider that allows the user to vary the
size of the bubbles. The server further provides an "active
bubbles" drop down menu that shows the bubbles and its radius that
are active. The server allows the user to add a sufficient number
of bubbles in the Whole Brain. The server also provides a "Delete
active bubbles" tool that enables the user to delete one or more
active bubbles within the boundaries of the Brain. The server
provides a "Next" tab to finalize the volume extraction as shown in
FIG. 13f.
The server provides a "continuous update" tool in a
three-dimensional window that enables to continuously update
contour evolution. The server further provides a "play" tab that
allows the user to play and pause Active Contour Evolution as shown
in FIG. 13f. The server further provides a "finish" tab that allows
the user to submit when the active contour evolution is done. The
server allows the user to change the "active label" to "clear
label" and edit the voxels when the active contour evolution goes
out of the boundaries of the Brain Parenchyma. The server allows
the user to edit the voxels by accessing the "brush" tool and
selecting appropriate brush and appropriate brush size. The server
allows the user to change the "active label" to "Whole Brain" and
add/edit the voxels when the active contour evolution has not
reached any part of the Brain Parenchyma. The server allows the
user to edit the voxels by accessing the "brush" tool and selecting
appropriate brush and appropriate brush size.
The server may render and save the Brain in at least one anatomical
pane and in three-dimensional format under the "Active Label" as
"Whole Brain" as shown in FIG. 13g. Once the segmentation and
volume extraction of the Whole Brain are complete, the server
enables the user to save the one or more first images, the
workspace, the mesh and the one or more segmented image with a
patient id name. The server enables the user to place the directory
location for the image files in the patient ID folder. The server
enables the user to capture screenshots of the segmented image at
all anatomical planes in such a way that Whole Brain is displayed
well with labels. The server further allows the user to capture
screenshots of the segmented image in a three-dimensional image
format with the Whole Brain zoomed in and displayed well.
Boundaries of the Whole Brain are shown in FIG. 13h. FIG. 13h shows
reference figures that illustrate the following. While segmenting
Whole brain, the arteries and sinuses should be removed from the
brain segmentation. The ventricles, (cerebrospinal fluid) csf
spaces, dura mater and skull are excluded. Sella turcica is removed
properly. The brainstem should include four image slices below the
cerebellum ends. The server also enables the user to check that the
whole brain parenchyma is included, using the "brush" tool if any
area is excluded or included. The server renders the segmented
image in the at least one anatomical plane and enables the user to
check for errors in the intensity of the image that might lead to
ring-like deficits appearing on the image after the bubbles are
evolved.
FIG. 14a-14c illustrate a process of segmentation of an
intracranial volume (ICV), according to one or more embodiments.
The process of segmentation of the ICV comprises the following
technical steps to be executed. While performing the segmentation,
a server enables a user to select "active label" as "ICV". In an
embodiment, RGB values assigned for the ICV are: R-126, G-84, and
B-126. The server further enables the user to select the
"Classification" tool under "pre-segmentation" tool as shown in
FIG. 14a and add a classifier "ICV". The classifier for "ICV"
comprises regions covering brain parenchyma, ventricles and csf
spaces. The server further enables the user to select the
"Classification" tool under the "pre-segmentation" tab and add a
classifier "Skull". The server renders a "speed image generation"
window and enables the user to check two options as shown in FIG.
14b under the "More" tool (shown in FIG. 14a). The two options
comprise (a) Include the intensities of voxels' neighbors as
aspects and (b) Include the voxels' coordinates as aspects. The two
options are configured to differentiate and train the classifiers.
The server provides a "Train Classifier" tool that allows the user
to train the classifier. Once the classifier is trained, the server
provides a "Next" tab to complete the training.
The server further provides a "Add bubbles at cursor" tool that
allows the user to add bubbles of appropriate size throughout the
trained area of the segmented image. The server allows the user to
evolve the bubbles until the bubbles cover the whole Intra Cranial
Cavity (ICV) properly. The server allows the user to change the
"active label" to "clear label" and edit the voxels when the active
contour evolution goes out of the boundaries of the ICV. The server
allows the user to edit the voxels by accessing a "brush" tool and
selecting appropriate brush and appropriate brush size. The server
allows the user to change the "active label" to "ICV" and edit/add
the voxels when the active contour evolution has not reached any
part of the ICV. The server allows the user to edit the voxels by
accessing the "brush" tool and selecting appropriate brush and
appropriate brush size. The server also provides a
"three-dimensional (3D) brush" to include the area of the ICV that
might have been missed after evolving the bubbles. The server also
allows the user to edit the missed area of the ICV in a sagittal
plane and to start the editing in mid sagittal slice. The server
also allows the user to use a two-dimensional brush instead of the
three-dimensional brush when an orbital area of the ICV is not
visible in the sagittal plane. The above steps are repeated for the
other side of the ICV.
Once the segmentation and volume extraction of the ICV are
complete, the server enables the user to save the one or more first
images, workspace, mesh and the one or more segmented images with a
patient id name. The server places the directory location for the
image files in the patient ID folder. Boundaries of the ICV in
different anatomical planes are shown in FIG. 14c. In an
embodiment, the boundaries of the ICV excludes dura mater, skull
and bones (Sella turcica area). The boundaries of the ICV include
arteries and sinuses in the segmentation. In an axial plane, the
boundaries of the ICV include four image slices below the
cerebellar end in the segmentation.
FIG. 15a-15d illustrate a process of segmentation of Cerebrum,
according to one or more embodiments. The process of segmentation
of the Cerebrum comprises the following technical steps. In an
embodiment, a server enables a user, via a user interface, to
upload the segmented image of a whole brain in Neuroimaging
Informatics Technology Initiative (NIfTI) format. The server allows
the user to change the "active label" as "clear label" and utilize
the "brush" tool to remove structures of Cerebellum and Brain stem
from the whole brain as shown in FIG. 15a. The server renders the
Cerebrum in at least one anatomical plane and a three-dimensional
format. The server enables the user to view the Cerebrum in the
three-dimensional format once the removal is done and use a cursor
and place it on a longitudinal fissure of the Cerebrum.
The server provides a "split" tool that enables the user to place a
line which traces the longitudinal fissure as shown in FIG. 15b.
The server shows the Cerebrum with an arrow pointing at one of the
left side and right side. The server allows the user to select the
"active label" as "Right Cerebral Hemisphere" when the arrow is
pointing at the left side. The RGB values for the Right Cerebral
Hemisphere are: R-107, G-101 and B-194. The server provides an
"accept" tab that allows the user to accept and update the
segmentation of 3D view format. The server further enables the user
to check in an axial plane whether the right Cerebral Hemisphere is
labelled properly.
The server allows the user to repeat the above steps starting from
utilizing the "split" tool for the left Cerebral Hemisphere. The
server enables the user to change the "active label" as "Left
Cerebral Hemisphere" as shown in FIG. 15c. The RGB values for the
left Cerebral Hemisphere are: R-0 G-181 B-121. The server allows
the user to select the "paint over" label as `whole brain`. The
server allows the user to use 3D brush and keep the size to its
maximum and use it over the left cerebrum which is present in the
whole brain label. Once the segmentation is done, the server allows
the user to save the segmented images of the Right Cerebrum and the
Left Cerebrum with the patient ID. The server also enables the user
to place the directory location for the image files in the patient
ID folder. The boundaries of the right cerebrum and the left
cerebrum are shown in FIG. 15d. FIG. 15d illustrates the following.
The boundaries of the Cerebrum are obtained by removing Brainstem
and Cerebellum from the whole brain segmentation. The boundaries of
the Cerebrum are further obtained by removing arteries and sinus
(if edits are seen). The segmented image rendered uses the
longitudinal fissure for separating the left cerebral hemispheres
and the right cerebral hemispheres. The Cerebrum includes the lobes
properly.
FIG. 16a-16c illustrate a process of segmentation of Cerebellum,
according to one or more embodiments. The process of segmentation
of the Cerebellum comprises the following technical steps. A server
enables a user, via a user interface, to upload a segmented image
of a whole brain in Neuroimaging Informatics Technology Initiative
(NIfTI) format. The server allows the user to change the "active
label" as "clear label" and utilize the "brush" tool to remove
structures of Cerebrum and Brain stem from the whole brain. The
server renders the Cerebellum in at least one of an anatomical
plane, and a three-dimensional format as shown in FIG. 16a. The
server enables the user to view the Cerebellum in the
three-dimensional format and use a cursor and place it on a
Vermis.
The server provides a "split" tool that enables the user to place a
line which traces mid of the Vermis. The server depicts the
Cerebellum with an arrow pointing at one of the left side and right
side. The server allows the user to select the "active label" as
"Right Cerebellar Hemisphere" as shown in FIG. 16b when the arrow
is pointing at the left side. The RGB values for the Right Cerebral
Hemisphere are: R-103, G-5 and B-173. The server provides an
"accept" tab that allows the user to accept and update the
segmentation of the Right Cerebral Hemisphere's 3D view format. The
server further enables the user to check in the axial plane whether
the right Cerebellar Hemisphere has been labelled properly.
The server allows the user to repeat the above steps starting from
utilizing the "split" tool for left Cerebellar Hemisphere or allows
the user to change the "active label" as "Left Cerebellar
Hemisphere". The RGB values are: R-0, G-145, B-16. The server
allows the user to select the "paint over" label as `whole brain`.
The server allows the user to use 3D brush and keep the size to its
maximum and use the 3d brush over the left cerebellum which is
present in the whole brain label. The server allows the user to
save the segmented images of the Right Cerebellum and the Left
Cerebellum with the patient ID. The server also enables the user to
place the directory location for the image files in the patient ID
folder. The boundaries of the right cerebellum and the left
cerebellum are shown in FIG. 16c. FIG. 16c illustrates the
following. The boundaries of the Cerebellum are obtained by
removing Brainstem and Cerebrum from the whole brain segmentation.
FIG. 16c depicts that in an axial plane, the segmented image uses
superior and middle cerebellar peduncle as a point of separation of
the Cerebellum from the Brainstem. FIG. 16d depicts that the
segmented image uses the transverse fissure for separating the
Cerebrum from the Cerebellum. The boundaries of the Cerebellum are
exactly obtained by removing transverse sinus, if not removed from
the segmented image of the Cerebellum.
FIG. 17a-17h illustrate a process of segmentation of Brainstem,
according to one or more embodiments. The process of segmentation
of the Brainstem comprises the following technical steps. A server
enables a user, via a user interface, to upload the image of the
whole brain in Neuroimaging Informatics Technology Initiative
(NIfTI) format. The server allows a user to change the "active
label" as "Brainstem". The RGB values are: R-0, G-216, B-249. The
server provides a "segmentation" tool that enables the user to set
a field of view (FOV) according to a region of interest (ROI) in
the anatomical planes as shown in FIG. 17a. The segmentation can be
done by any of (a) classification, (b) thresholding, (c)
clustering, and (d) edge attraction. The server provides a "speed
image generation" window upon selecting the "thresholding" tool as
shown in FIG. 17b. The "speed image generation" window provides an
"upper threshold" slider and a "lower threshold" slider. The server
enables the user to vary an upper threshold value and a lower
threshold value using the "upper threshold" slider and the "lower
threshold" slider, respectively.
In an embodiment, the upper threshold value is varied so that the
upper threshold value is moved to the highest value possible and
the lower threshold value is varied to increase the lower threshold
value slowly till the csf is removed from the classifier or the
overlay as shown in FIG. 17c. The server renders volume of the
Brainstem as per the threshold values, which enables the user to
check that the voxels of the Brainstem are included in the overlay
area. The server provides a "Next" tab that is to be clicked when
the threshold values are adjusted and finalized. The server further
provides a "add bubbles on cursor" tool that allows the user to add
bubbles of appropriate size within the boundaries of the brainstem
as shown in FIG. 17d.
The server allows the user to evolve the bubbles until the bubbles
cover the brainstem entirely as shown in FIG. 17e. The server
further provides a "Finish" tab that enables the user to click when
the bubbles are populated in the Brainstem. The server further
provides a "Brush" tool that allows the user to edit/add or delete
voxels of the brainstem when the bubbles are overestimated or not
reached the entire structure of the Brainstem, respectively. Once
the segmentation and volume extraction of the Brainstem are
complete, the server renders the Brainstem in at least one
anatomical plane and the three-dimensional format as shown in FIG.
17f. The server enables the user to save the one or more first
images, the workspace, the mesh and the one or more segmented
images with a patient id name. The server enables the user to place
the directory location for the image files in the patient ID
folder. The final volumes of the Brainstem are rendered in the at
least one anatomical plane and in a three-dimensional format view
as shown in FIG. 17f.
FIGS. 17g, and 17h depict boundaries of the Brainstem. The
Brainstem is separated by a thin hyperintense boundary from a
Thalamus in a superior aspect. Middle cerebellar peduncle and
superior cerebellar peduncle separate the Brainstem from a
Cerebellum. The segmented image does not include the tectal plate
in the segmentation. The image includes 4 slices below the
cerebellum end.
FIG. 18 illustrates a process of segmentation of Midbrain,
according to one or more embodiments. The process of segmentation
of the Midbrain comprises the following technical steps. A server
allows a user to upload the main file on ITK snap. The server
provides an "add segmentation file" tool that enables the user to
Upload Brainstem segmentation files. The server further provides a
"brush" tool and "3D brush" tool under the "brush" tool that
enables the user to remove Pons and Medulla oblongata from the
segmentation. The server further enables to select "active label"
as "clear label" before removing the Pons and the Medulla Oblongata
from the Brainstem segmentation file.
The server renders the segmented image of the Midbrain in the at
least one anatomical plane and a three-dimensional format and
enables the user to check the boundaries of the midbrain in the
three anatomical planes. The server allows the user to select the
"active label" as "Midbrain" and "paint over" label as "Brainstem".
In an embodiment, RGB values are: R-255, G-130, B-153. The server
further enables the user to paint the Midbrain region using a
"brush" tool as 3D brush and add/edit voxels within the
brainstem.
Once the segmentation and volume extraction of the Midbrain are
complete, the server enables the user to save the one or more first
images, workspace, mesh and the one or more segmented images with a
patient id name. The server enables the user to place a directory
location for the image files in a patient ID folder. Boundaries of
the Midbrain are shown in FIG. 18. FIG. 18 shows that Pons is
separated from the Midbrain by Superior Pontine Sulci. FIG. 18
further shows that the midbrain is superiorly separated from
Thalamus by a thin hyperintense border.
FIG. 19 illustrates a process of segmentation of Pons, according to
one or more embodiments. The process of segmentation of the Pons
comprises the following technical steps. A server allows a user to
upload the main file on ITK snap. The server provides an "add
segmentation file" that enables the user to Upload Brainstem
segmentation file. The server further provides a "brush" tool and
"3D brush" tool under "brush" tool that enables the user to remove
Midbrain and Medulla oblongata from the segmentation. The server
further provides an "active label" as "clear label" before removing
the Pons and the Medulla Oblongata from the Brainstem segmentation
file.
The server renders the segmented image of the Pons in the at least
one anatomical plane and a three-dimensional format and enables the
user to check the boundaries of the midbrain in the three
anatomical planes. The server allows the user to select the "active
label" as "Pons" and "paint over" label as "Brainstem". In an
embodiment, RGB values are: R-255, G-182, B-193. The server further
enables the user to paint the Pons region using a "brush" tool as
"3D brush".
Once the segmentation and volume extraction of the Pons are
complete, the server enables the user to save the one or more first
images, the workspace, the mesh and the one or more segmented
images with patient id name. The server enables the user to place
the directory location for the image files in the patient ID
folder. Boundaries of the Pons are shown in FIG. 19. FIG. 19 shows
that Pons is separated from the Midbrain by a Superior Pontine
Sulci. FIG. 19 further shows that the Pons is separated from
Medulla by an inferior pontine sulcus.
FIGS. 20a-20e illustrate a process of segmentation of Amygdala,
according to one or more embodiments. The process of segmentation
of the Amygdala comprises the following technical steps. One or
more first images of a region of interest (i.e., skull) are
uploaded to an ITK snap layer of a server. A label file is imported
for the Amygdala. Amygdala label file comprises predefined RGB
values. In an embodiment, the predefined RGB values of a right
Amygdala assigned are R-255, G-94, B-97. In another embodiment, the
predefined RGB values of a left Amygdala assigned are R-253, G-255,
B-89. A server enables the user to access a contrast inspector
drop-down tab via a user interface to adjust the contrast so that
grey matter (GM) and white matter (WM) differentiation is optimum
as shown in FIG. 20a. The one or more first images are rendered in
at least one anatomical plane such as a sagittal plane, an axial
plane, and a coronal plane to readily enable a user to visualize
the Amygdala in the at least one anatomical plane and identify a
location, position and shape of the Amygdala. The Amygdala
comprises a right Amygdala and a left Amygdala. The server enables
the user to select "active label" as "right Amygdala" or "left
Amygdala" accordingly.
The Amygdala can be identified by the following. The Amygdala is an
ovoid mass of gray matter situated in the superomedial portion of
the temporal lobe, partly above the tip of the inferior horn of the
lateral Ventricle. The Amygdala occupies the superior part of the
anterior segment of the uncus and partially overlies the head of
the Hippocampus, being separated from that structure by the uncal
recess of the inferior horn of the Lateral Ventricle. On the
superomedial surface of the uncus, the Amygdala forms a distinct
protrusion, the Semilunar Gyms, which corresponds to the Cortical
Amygdaloid Nucleus. It is separated from the Ambient Gyms by the
Semiannual or Amygdaloid Sulcus, which forms the boundary between
the Amygdala and the Entorhinal Cortex.
Upon determining, when the one or more first images do not comprise
optimized quality in terms of shape, boundary and volume of at
least one structure, the images can be further manually edited by
performing manual segmentation. The server enables the user to move
to an image slice using a sagittal plane when the right Amygdala
disappears and scrolls to the next visible image slice. The server
provides a "polygon" tool on a main toolbar as shown in FIG. 20b.
The "polygon" tool, upon selecting, enables the user to perform the
manual segmentation by drawing and filling polygons in orthogonal
image slices. In an embodiment, the manual segmentation can be done
individually in the at least one anatomical plane. The manual
segmentation, via the polygon tool, enables the user to add points
to the polygon and edit the completed polygon.
The "polygon" tool enables the user to zoom in and out (hold and
drag) to view any specific portion of the Amygdala. The "polygon"
tool further enables the user to place and move 3D cursor, scroll
through image slices and scroll through image components to view,
edit and correct the volume, shape and structure of the Amygdala.
The server further provides an "active label" tool under
"segmentation label". Under the "active label", the user is enabled
to select an appropriate label (i.e., right HC in this instance) as
shown in FIG. 20c. The server further enables the user to select
the "paint over" tool as all labels. The server rendered enables
the user to choose opacity so as not to obscure/hide tissue
boundaries. In an embodiment, the opacity ranges between 15-30. The
server enables the user to outline the outermost border of the
Amygdala using the image slice chosen in the at least one
anatomical plane as shown in FIG. 20d. In an embodiment, a first
color (e.g., pink) is used for an active polygon and a second color
(e.g., red) stands for a completed polygon.
The server further enables the user to retrace borders of the
Amygdala and detect any missing pixels or voxels by zooming in. The
server further provides a "brush" tool. The "brush" tool further
enables the user to edit and add the missing pixels/voxels by
selecting an appropriate brush (e.g., round brush) and appropriate
bush size. If the edits have been done more than the actual
voxels/pixels (i.e., in case of over estimation), the server
enables the user to select the "active label" as "clear label" and
edit the voxels.
The server also enables the user to segment left Amygdala and
extract volumes of the left Amygdala. Once the right Amygdala is
completed repeat the above steps for the left Amygdala. To proceed
with the left Amygdala, the user should change the "active label"
as left Amygdala before starting left Amygdala. The segmented image
and extracted volumes of both the left Amygdala and the right
Amygdala are shown in FIG. 20e.
Once the segmentation and volume extraction of the right Amygdala
and left Amygdala are complete, the server enables the user to save
the one or more first images, the workspace, the mesh and the one
or more segmented image with patient id name. The server enables
the user to place the directory location for the image files in the
patient ID folder. The server enables the user to capture
screenshots of the segmented image at all anatomical planes in such
a way that both the Left Amygdala and Right Amygdala are displayed
well with labels. The server further allows the user to capture
screenshots of the segmented image in a three-dimensional image
format with both the left Amygdala and right Amygdala zoomed
well.
Boundaries of the segmented Amygdala illustrates the following: The
Amygdala lies in an anterior aspect of the Hippocampus. The
Amygdala is best viewed in the sagittal plane and axial plane.
Sulci lines and temporal horn of the Ventricle are useful while
segmenting the Amygdala. The lateral aspect of the Hippocampus is
differentiated from the White matter. The posterior aspect is
separated from the Hippocampal head and the Fimbria. However,
medially, the Hippocampal head and Amygdala seem to attach or have
no space in between. This attachment area should be segmented by
viewing the thin Hyperintense border of the Hippocampal head.
FIG. 21a-21g illustrate a process of segmentation of Basal Ganglia,
according to one or more embodiments. The process of segmentation
of the Basal Ganglia comprises the following technical steps. One
or more first images of a region of interest (i.e., skull) are
uploaded to an ITK snap layer of a server. A label file is imported
for the Basal Ganglia. Basal Ganglia label file comprises
predefined RGB values. In an embodiment, the predefined RGB values
of the Basal Ganglia assigned are R-122, G-180, B-181. A server
enables the user to access a contrast inspector drop-down tool via
a user interface to adjust the contrast so that grey matter (GM)
and white matter (WM) differentiation is optimum as shown in FIG.
21a. The one or more first images comprising the Basal Ganglia are
rendered in at least one anatomical plane such as a sagittal plane,
an axial plane, and a coronal plane to readily enable a user to
visualize the Basal Ganglia in the at least one anatomical plane
and identify a location, a position and a shape of the Basal
Ganglia. The Basal Ganglia comprises a right Basal Ganglia and a
left Basal Ganglia.
The boundaries of the Basal Ganglia can be identified considering
the following: The tracing of Caudate Nucleus starts at a section
where it is first visualized in the Frontal Horn of Lateral
Ventricles and ends at the section where it is no longer
identifiable; the Nucleus Accumbens is used as Ventral boundary,
the Lateral ventricle is used as Medial boundary, and Internal
Capsule is used as Lateral Boundary. For the boundary of the
Putamen, the medial boundary is the Internal capsule (Anterior
Putamen) and the External Pallidum (Posterior Putamen); the lateral
boundary is defined by the External Capsule.
Upon determining, when the one or more segmented images do not
comprise optimized quality in terms of shape, boundary and volume
of at least one structure, the segmented images can be further
manually edited by performing manual segmentation. The server
enables the user to move to an image slice using a sagittal plane
when right Basal Ganglia disappears and moves to the next visible
image slice. The server provides a "polygon" tool on a main toolbar
as shown in FIG. 21b. The "polygon" tool, upon selecting, enables
the user to perform the manual segmentation by drawing and filling
polygons in orthogonal image slices. In an embodiment, the manual
segmentation can be done individually in each anatomical plane. The
manual segmentation, via the polygon tool, enables the user to add
points to the polygon and edit the completed polygon.
The "polygon" tool enables the user to zoom in and out (hold and
drag) to view any specific portion of the Basal Ganglia. The
"polygon" tool further enables the user to place and move 3D
cursor, scroll through image slices and scroll through image
components to view, edit and correct the volume, shape and
structure of the Basal Ganglia. The server further provides an
"active label" tool under "segmentation label". Under the "active
label", the user is enabled to select an appropriate label (i.e.,
right HC in this instance) as shown in FIG. 21c. The server further
enables the user to select the "paint over" tool as all labels. The
server enables the user to choose opacity so as not to obscure/hide
tissue boundaries. In an embodiment, the opacity ranges between
15-30. The server enables the user to outline the outermost border
of the Basal Ganglia using the image slice chosen as shown in FIGS.
21d and 21e. In an embodiment, a first color (e.g., pink) is used
for an active polygon and a second color (e.g., red) stands for
completed polygon.
The server further enables the user to retrace borders of the Basal
Ganglia and detect any missing pixels or voxels by zooming in. The
server further provides a "brush" tool. The "brush" tool further
enables the user to edit and add the missing pixels by selecting an
appropriate brush (e.g., round brush) and appropriate bush size. If
the edits have been done more than the actual voxels (i.e., in case
of over estimation), the server enables the user to select the
"active label" as "clear label" and edit the voxels.
The server also enables the user to segment left Basal Ganglia and
extract volumes of the left Basal Ganglia. Once the right Basal
Ganglia is completed repeat the above steps for the left Basal
Ganglia. To proceed with the left Basal Ganglia, the user should
change the "active label" as "left HC" before starting the left
Basal Ganglia. The segmented image and extracted volumes of both
the left Basal Ganglia and the right Basal Ganglia are shown in
FIG. 21f.
Once the segmentation and volume extraction of the right Basal
Ganglia and left Basal Ganglia are complete, the server enables the
user to save the one or more first images, the workspace, the mesh
and the one or more segmented image with patient id name. The
server enables the user to place the directory location for the
image files in the patient ID folder. The server enables the user
to capture screenshots of the segmented image at all anatomical
planes in such a way that both the Left Basal Ganglia and Right
Basal Ganglia are displayed well with labels. The server further
allows the user to capture screenshots of the segmented image in a
three-dimensional image format with both the left Basal Ganglia and
right Basal Ganglia zoomed well. Boundaries of the right Basal
Ganglia and the left Basal Ganglia are shown in FIG. 21g. The
server further enables to segment the Caudate Nucleus by removing
Putamen from the Basal Ganglia segmentation and adding bubbles to
the Caudate Nucleus with its label. The RGB values are: R-104,
G-176, B-138.
FIG. 21g illustrates the following. The Basal Ganglia comprises:
the striatum; both dorsal striatum (Caudate Nucleus and Putamen)
and Ventral Striatum (Nucleus Accumbens and Olfactory Tubercle),
Globus Pallidus, Ventral Pallidum, Substantia Nigra and Subthalamic
Nucleus. In this segmentation, only the Caudate Nucleus and the
Putamen are included. The Subthalamic Nucleus and the Substantia
Nigra are segmented separately. The Caudate Nucleus is a C-shaped
structure that is associated with the lateral wall of Lateral
Ventricle. Caudate is the largest at its anterior pole (the head),
and its size diminishes posteriorly as it follows the course of the
Lateral Ventricle (the body) all the way to the Temporal Lobe (the
tail), where it terminates at the Amygdaloid Nuclei. The Putamen is
separated from the Caudate Nucleus by the Anterior limb of the
internal capsule. The Putamen is connected to the Caudate head by
bridges of cells that cut across the Internal Capsule.
FIG. 22a-22f illustrate a process of segmentation of Thalamus,
according to one or more embodiments. The process of segmentation
of the Thalamus comprises the following technical steps. One or
more first images of the region of interest (i.e., skull) are
uploaded to an ITK snap layer of a server. An "active label" is
selected as "Thalamus". Thalamus label file comprises predefined
RGB values. In an embodiment, the predefined RGB values of the
Thalamus are R-247, G222, B-130. A server enables a user to access
a contrast inspector drop-down tool via a user interface to adjust
the contrast so that grey matter (GM) and white matter (WM)
differentiation is optimum as shown in FIG. 22a. The one or more
first images comprising Thalamus are rendered in at least one
anatomical plane (such as a sagittal plane, an axial plane, and a
coronal plane). The one or more first images readily enable a user
to visualize the Thalamus in the at least one anatomical plane and
identify a location, a position and a shape of the Thalamus. The
Thalamus comprises a right Thalamus and a left Thalamus.
Upon determining, when the one or more segmented images do not
comprise optimized quality in terms of shape, boundary and volume
of at least one structure, the segmented images can be further
manually edited by performing manual segmentation. The server
enables the user to move to an image slice using a sagittal plane
when right Thalamus disappears and moves to the next visible image
slice. The server provides a "polygon" tool on a main toolbar as
shown in FIG. 22b. The "polygon" tool, as shown in FIG. 22b, upon
selecting, enables the user to perform the manual segmentation by
drawing and filling polygons in orthogonal image slices. In an
embodiment, the manual segmentation can be done individually in the
anatomical plane. The manual segmentation, via the polygon tool,
enables the user to add points to the polygon and edit the
completed polygon.
The "polygon" tool enables the user to zoom in and out (hold and
drag) to view any specific portion of the Thalamus. The "polygon"
tool further enables the user to place and move 3D cursor, scroll
through image slices and scroll through image components to view,
edit and correct the volume, shape and structure of the Thalamus.
The server further provides an "active label" tool under
"segmentation label". Under the "active label", the user is enabled
to select an appropriate label (i.e., right Thalamus in this
instance). The server further enables the user to select a "paint
over" tool as all labels. The server rendered enables the user to
choose opacity so as not to obscure/hide tissue boundaries. In an
embodiment, the opacity ranges between 15-30. The server enables
the user to outline the outermost border of the Thalamus using the
image slice chosen as shown in FIGS. 22c and 22d. In an embodiment,
a first color (e.g., pink) is used for active polygon and a second
color (e.g., red) stands for completed polygon.
The server further enables the user to retrace borders of the
Thalamus and detect any missing pixels or voxels by zooming in. The
server further provides a "brush" tool. The "brush" tool further
enables the user to edit and add the missing pixels by selecting an
appropriate brush (e.g., round brush) and appropriate bush size. If
the edits have been done more than the actual voxels (i.e., in case
of over estimation), the server enables the user to select the
"active label" as "clear label" and edit the voxels.
The server also enables the user to segment left Thalamus and
extract volumes of the left Thalamus. Once the right Thalamus is
completed repeat the above steps for the left Thalamus. To proceed
with the left Thalamus, the user should change the "active label"
as "left Thalamus" before starting left Thalamus. The segmented
image and extracted volumes of both the left Thalamus and the right
Thalamus are shown in FIG. 22e.
Once the segmentation and volume extraction of the right Thalamus
and left Thalamus are complete, the server enables the user to save
the one or more first images, the workspace, the mesh and the one
or more segmented image with a patient id name. The server further
enables the user to place the directory location for the image
files in the patient ID folder. The server enables the user to
capture screenshots of the segmented image at all anatomical planes
in such a way that both the Left Thalamus and Right Thalamus are
displayed well with labels. The server further allows the user to
capture screenshots of the segmented image in a three-dimensional
image format with both the left Thalamus and right Thalamus zoomed
well. Boundaries of the right Thalamus and the left Thalamus are
shown in FIG. 22f.
FIG. 22f illustrates the boundaries of the Thalamus. Anteriorly,
the Thalamus is defined by the posterior boundary of
Interventricular Foramen, a channel allowing the movement of
Cerebrospinal Fluid from the lateral to the Third Ventricles.
Posteriorly, it is defined by an expansion called the Pulvinar.
Inferior border: The Tegmentum or floor of the Midbrain.
Medially--lateral wall of the third Ventricle.
FIGS. 23a-23c illustrate a process of segmentation of Substantia
Nigra, according to one or more embodiments. The process of
segmentation of the Substantia Nigra comprises the following
technical steps. One or more first images of the region of interest
(i.e., skull) are uploaded to an ITK snap layer of a server. A
label file is imported for the Substantia Nigra. Substantia Nigra
label file comprises predefined RGB values. In an embodiment, the
predefined RGB values of the Substantia Nigra assigned are R-255,
G-187, B-188. A server enables a user to access a contrast
inspector drop-down tool via a user interface to adjust the
contrast so that grey matter (GM) and white matter (WM)
differentiation is optimum as shown in FIG. 23a. The one or more
first images comprising the Substantia Nigra are rendered in at
least one anatomical plane such as a sagittal plane, an axial
plane, and a coronal plane. The server readily enables a user to
visualize the Substantia Nigra in the at least one anatomical plane
and identify a location, a position, and a shape of the Substantia
Nigra.
In an embodiment, upon determining, when the one or more segmented
images do not comprise optimized quality in terms of shape,
boundary and volume of at least one structure, the segmented images
can be further manually edited by performing manual segmentation.
The server provides a "polygon" tool on a main toolbar as shown in
FIG. 23b. The "polygon" tool, upon selecting, enables the user to
perform the manual segmentation by drawing and filling polygons in
orthogonal image slices. In an embodiment, the manual segmentation
can be done individually in the anatomical plane. The manual
segmentation, via the polygon tool, enables the user to add points
to the polygon and edit the completed polygon.
The "polygon" tool enables the user to zoom in and out (hold and
drag) to view any specific portion of the Substantia Nigra. The
"polygon" tool further enables the user to use a "cursor chase" tab
and place and move 3D cursor, scroll through image slices and
scroll through image components to view, edit and correct the
volume, shape and structure of the Substantia Nigra. The server
further provides an "active label" tool under "segmentation label".
Under the "active label", the user is enabled to select an
appropriate label (i.e., Substantia Nigra in this instance). The
server further enables the user to select the "paint over" tool as
all labels. The server, via the user interface rendered, enables
the user to choose opacity so as not to obscure/hide tissue
boundaries. In an embodiment, the opacity ranges between 15-30. The
server enables the user to outline the outermost border of the
Substantia Nigra using the image slice chosen. In an embodiment, a
first color (e.g., pink) is used for active polygon and a second
color (e.g., red) stands for completed polygon.
The server further enables the user to retrace borders of the
Substantia Nigra and detect any missing pixels or voxels by zooming
in. The server further provides a "brush" tool. The "brush" tool
further enables the user to edit and add the missing pixels by
selecting an appropriate brush (e.g., round brush) and appropriate
bush size. If the edits have been done more than the actual voxels
(i.e., in case of over estimation), the server enables the user to
select the "active label" as "clear label" and edit the voxels.
Once the segmentation and volume extraction of the Substantia Nigra
and the left Substantia Nigra are complete, the server renders the
Substantia Nigra as shown in FIG. 23c and enables the user to save
the one or more first images, the workspace, the mesh and the one
or more segmented images with patient id name. The server enables
the user to place the directory location for the image files in the
patient ID folder. The server enables the user to capture
screenshots of the segmented image at all anatomical planes in such
a way that the Substantia Nigra is displayed well with labels. The
server further allows the user to capture screenshots of the
segmented image in a three-dimensional image format with the
Substantia Nigra zoomed well.
FIGS. 24a-24j illustrate a process of segmentation of Frontal
Lobes, according to one or more embodiments. The process of
segmentation of the Frontal Lobes comprises the following technical
steps. While performing the segmentation, a server enables a user
to select "active label" as "Frontal Lobes". In an embodiment, RGB
Values assigned for the Frontal Lobes are: R-0, G-241, and B-193.
The server further enables the user to select "Contour Segmentation
mode" i.e., Semi-automatic segmentation. The contour segmentation
allows the user to select semi-automatic active contour
segmentation and start the semi-automatic segmentation as shown in
FIG. 24a. The contour segmentation enables the user to adjust the
boundaries of a region of interest covering the entire brain. The
server enables a user to assign an "active label" as "Frontal
Lobes". One or more first images comprising the Frontal Lobes are
rendered in at least one anatomical plane and a three-dimensional
format. The server enables the user to identify the Frontal Lobes
in the at least one anatomical plane.
Once the Frontal Lobes is identified, the server enables the user
to mark a region of interest covering the Frontal Lobes and check
whether the Frontal Lobes is covered in the anatomical planes as
shown in FIG. 24b. The server provides a "segment 3D" tool that
allows the user to start semi-automatic segmentation. The user upon
clicking the "segment 3D" tool, a toolbar appears. The toolbar
provides a "pre-segmentation" tool. The "pre-segmentation" tool
provides a "thresholding", "classification", "clustering", and
"edge attraction". The user can select any of the four tools. For
example, the "classification" tool is selected by the user.
Before starting the "classification" type segmentation, the server
enables the user to choose a label indicating "Frontal Lobes" under
the "Segmentation labels" tool. The "active label" under the
"Segmentation label" tool is used to record and save information
(e.g., volumes, boundaries, manual edits performed to the
segmentation, etc.) obtained as a result of the segmentation
performed on the at least one structure i.e., Frontal Lobes in this
case. The server further provides a "brush" tool that allows the
user to select appropriate brush (e.g., round brush) and
appropriate brush size to mark the brain structures (e.g., GM, WM,
frontal cortex). The Brain structure marked comprises brain
structures (e.g., GM, WM, frontal cortex). The "classification"
tool allows the user to classify between Frontal lobes and
Intracranial Volume (ICV) by providing two labels "Frontal lobes"
and "ICV". The "Frontal lobes" label is used to classify between
white matter and grey matter. The "ICV" label is used to classify
between dura, skull bone, ventricles or csf. In an embodiment, if
there is an error, the "classification" tool further allows the
user to add a third label as "Ventricles" to classify the
Ventricles separately. The different tissue samples comprise white
matter (WM) and grey matter (GM).
The server allows the user to mark tissue samples such as the WM
and GM. The server further provides a "train classifier" tool that
allows the user to train the classifier by clicking on the "train
classifier" tool. The server further renders a speed image that
shows the classification as shown in FIG. 24c. The "train
classifier" assigns a first probability value to a voxel belonging
to the "foreground" class and a second probability value to a voxel
belonging to all other classes. Once the classifier is trained
using the manual segmentation (i.e., marking and differentiating
the tissue samples, Frontal Lobes, ICV, and the Ventricles) the
classifier automatically segments at least one structure within the
one or more first images of a different patient in future based on
micro-ethnicity information, age and gender. The server provides a
"Next" tab to complete the segmentation process.
The server provides a "Add bubble at cursor" tool as shown in FIG.
24d that allows the user to populate bubbles of appropriate sizes
exactly in the Frontal Lobes in at least three anatomical planes.
Further the server provides a "Bubble radius" slider that allows
the user to vary the size of the bubbles. The server further
provides an "active bubble" drop down menu that shows the bubbles
and its radius that are active. The server allows the user to add a
sufficient number of bubbles in the Frontal Lobes. The server
provides a "Next" tab to finalize the volume extraction.
The server provides a "continuous update" tool in a
three-dimensional window that enables to continuously update
contour evolution. The server further provides a "play" tab that
allows the user to play and pause Active Contour Evolution as shown
in FIG. 24e. The server further provides a "finish" tab that allows
the user to submit when the active contour evolution is done. In an
embodiment, the server allows the user to change the "active label"
to "clear label" and edit the voxels when the active contour
evolution goes out of the boundaries of the Frontal Lobes. The
server allows the user to edit the voxels by accessing the "brush"
tool and selecting appropriate brush and appropriate brush size. In
another embodiment, the server allows the user to change the
"active label" to "Frontal Lobes" and edit/add the voxels when the
active contour evolution has not reached any part of Frontal
Cortex. The server allows the user to edit the voxels by accessing
the "brush" tool and selecting appropriate brush and appropriate
brush size.
The server may render and save the Frontal Lobes in at least one
anatomical pane and in three-dimensional format under the "Active
Label" as "Frontal Lobes". Once the segmentation and volume
extraction of the Frontal Lobes are complete, the server enables
the user to save the one or more first images, the workspace, the
mesh and the one or more segmented image with patient id name. The
server enables the user to place the directory location for the
image files in the patient ID folder. The server enables the user
to capture screenshots of the segmented image at all anatomical
planes in such a way that Frontal Lobes are displayed well with
labels. The server further allows the user to capture screenshots
of the segmented image in a three-dimensional image format with the
Frontal Lobes zoomed in and displayed well.
The server renders the Frontal lobes in a three-dimensional format
as shown in FIG. 24f The server depicts the Frontal lobes with an
arrow pointing at one of the left side, and right side as shown in
FIG. 24g. The server allows the user to select the "active label"
as "Left Frontal Lobes" when the arrow is pointing at the right
side as shown in FIG. 24h. Similarly, the server allows the user to
select the "active label" as "Right Frontal Lobes" when the arrow
is pointing at the left side. The RGB values for the Left Frontal
Lobes are: R-226, G-147, B-90. The server provides an "accept" tab
that allows the user to accept and update the segmentation of 3D
view format. The server further enables the user to check in an
axial plane whether the Left Frontal Lobes has been labelled
properly. The server provides a cursor and places it on a
longitudinal fissure. The server provides a "split" tool that
enables the user to place a line which traces the longitudinal
fissure. The server further renders an arrow on the
three-dimensional window. The server then renders the Frontal Lobes
in at least one of three-dimensional format and at least one
anatomical plane as shown in FIG. 24i.
FIG. 24j comprise reference figures that illustrate the boundaries
of the Frontal Lobes. Central Sulcus identifies the posterior
border of the Frontal Lobe. The Sylvian Fissure demarcates the
inferior border of the Frontal Lobe. The superior and middle
Frontal Gyri are divided by the Superior Frontal Sulcus. The middle
and inferior Frontal Gyri are divided by the Inferior Frontal
Sulcus. Do not include the Corpus Callosum and the Basal
Ganglia.
FIGS. 25a-25i illustrate a process of segmentation of Parietal
Lobes, according to one or more embodiments. The process of
segmentation of the Parietal Lobes comprises the following
technical steps. While performing the segmentation, a server
enables a user to select "active label" as "Parietal Lobes". In an
embodiment, RGB Values assigned for the Ventricles are: R-194,
G-255, and B-187. The server further enables the user to select
"Contour Segmentation mode" i.e., Semi-automatic segmentation. The
contour segmentation allows the user to select semi-automatic
active contour segmentation and start the semi-automatic
segmentation as shown in FIG. 25a. The contour segmentation enables
the user to adjust the boundaries of the region of interest
covering the Parietal Lobes. Once the `active label` is assigned as
"Parietal Lobes", one or more first images are rendered in at least
one anatomical plane and a three-dimensional format. The server
enables the user to identify the Parietal Lobes in the at least one
anatomical plane.
Once the Parietal Lobes is identified, the server enables the user
to mark a region of interest covering the Parietal Lobes and check
whether the Parietal Lobes is covered in the anatomical planes as
shown in FIG. 25b. The server provides a "segment 3D" tool that
allows the user to start semi-automatic segmentation. The user upon
clicking the "segment 3D" tool a toolbar appears. The toolbar
provides a "pre-segmentation" tool. The "pre-segmentation" tool
provides a "thresholding", "classification", "clustering", and
"edge attraction". The user can select any of the four tools. For
example, the "classification" tool is selected by the user.
Before starting the "classification" type segmentation, the server
enables the user to choose a label indicating "Parietal Lobes"
under the "Segmentation labels" tool. The "Segmentation label" tool
is used to record and save information (e.g., volumes, boundaries,
manual edits performed to the segmentation, etc.) obtained as a
result of the segmentation performed on the at least one structure
i.e., Parietal Lobes in this case. The server further provides a
"brush" tool that allows the user to select appropriate brush
(e.g., round brush) and appropriate brush size to mark the brain
structures (e.g., GM, WM, frontal cortex) as shown in FIG. 25c
under the "Parietal Lobes" label of the "Segmentation labels". The
Brain structure marked comprises brain structures (e.g., GM, WM,
frontal cortex). The "classification" tool allows the user to
classify between Parietal lobes and Intracranial Volume (ICV) by
providing two labels "Parietal lobes" and "ICV". The "Right
Parietal lobes" label is used to classify between white matter and
grey matter. The "ICV" label is used to classify between dura,
skull bone, ventricles or csf. In an embodiment, if there is an
error, the "classification" tool further allows the user to add a
third label as "Ventricles" to classify the Ventricles separately.
The different tissue samples comprise white matter (WM) and grey
matter (GM).
The server allows the user to mark tissue samples such as the WM
and GM. The server further provides a "train classifier" tool that
allows the user to train the classifier by clicking on the "train
classifier" tool. The server further renders a speed image that
shows the classification. The "train classifier" assigns a
probability value to a voxel belonging to the "foreground" class
vs. belonging to all other classes. Once the classifier is trained
using the manual segmentation (i.e., marking and differentiating
the tissue samples, Parietal Lobes, ICV, and the Ventricles) the
classifier automatically segments at least one structure within the
one or more first images of a different patient in future based on
micro-ethnicity information, age and gender. The server provides a
"Next" tab to complete the segmentation process.
The server provides a "Add bubble at cursor" tool as shown in FIG.
25d that allows the user to populate bubbles of appropriate sizes
exactly in the Parietal Lobes in at least three anatomical planes
to exactly extract the volume of the Right Parietal Lobes as shown
in FIG. 25e. Further the server provides a "Bubble radius" slider
that allows the user to vary the size of the bubbles. The server
further provides an "active bubbles" drop down menu that shows the
bubbles and its radius that are active. The server allows the user
to add a sufficient number of bubbles in the Parietal Lobes. The
server provides a "Next" tab to finalize the volume extraction.
The server provides a "continuous update" that enables it to
continuously update contour evolution. The server further provides
a "play" tab that allows the user to run, play, and pause Active
Contour Evolution as shown in FIG. 25f. The server further provides
a "finish" tab that allows the user to submit when the active
contour evolution is done. In an embodiment, the server allows the
user to change the "active label" to "clear label" and edit the
voxels when the active contour evolution goes out of the boundaries
of the Parietal Lobes. The server allows the user to edit the
voxels by accessing the "brush" tool and selecting appropriate
brush and appropriate brush size. In another embodiment, the server
allows the user to change the "active label" to "Parietal Lobes"
and edit the voxels when the active contour evolution has not
reached any part of the Right Parietal Lobes. The server allows the
user to edit the voxels by accessing the "brush" tool and selecting
appropriate brush and appropriate brush size.
The server may render and save the Parietal Lobes in at least one
anatomical pane and in three-dimensional format under the "Active
Label" as "Right Parietal Lobes". Once the segmentation and volume
extraction of the Right Parietal Lobes are complete, the server
enables a user to save the one or more first images, the workspace,
the mesh and the one or more segmented image with patient id name.
The server enables the user to place the directory location for the
image files in the patient ID folder. The server enables the user
to capture screenshots of the segmented image at all anatomical
planes in such a way that Parietal Lobes are displayed well with
labels. The server further allows the user to capture screenshots
of the segmented image in a three-dimensional image format with the
Parietal Lobes zoomed in and displayed well.
The server renders the Parietal lobes in a three-dimensional
format. The server depicts the Parietal lobes with an arrow
pointing at one of the left side, and right side as shown in FIG.
25g. The server provides a cursor and places it on a longitudinal
fissure. The server provides a "split" tool that enables the user
to place a line which traces the longitudinal fissure. The server
further renders an arrow on the three-dimensional window. The
server allows the user to select the "active label" as "Left
Parietal Lobes" when the arrow is pointing at the right side as
shown in FIG. 25h. The server allows the user to select the "active
label" as "Right Parietal Lobes" when the arrow is pointing at the
left side. The server allows the user to select the "active label"
as "Right Parietal Lobes" when the arrow is pointing at the left
side. The RGB values for the Left Parietal Lobes are: R-252, G-0,
B-157. The server provides an "accept" tab that allows the user to
accept and update the segmentation of 3D view format. The server
further enables the user to Check in the Axial plane whether the
Left Parietal Lobes has been labelled properly.
FIG. 25i are reference figures that illustrate the boundaries of
the parietal Lobes. Central Sulcus separates the parietal lobe from
the frontal lobe. Parieto-occipital sulcus separates the Parietal
and Occipital lobes. Lateral Sulcus (Sylvian Fissure) is the most
lateral boundary, separating it from the Temporal Lobe. The
Longitudinal fissure divides the two hemispheres. The Parietal
Lobes do not include Corpus Callosum.
FIG. 26a-26h illustrate a process of segmentation of Occipital
Lobes, according to one or more embodiments. The process of
segmentation of the Occipital Lobes comprises the following
technical steps. While performing the segmentation, a server
enables a user to select "active label" as "Occipital Lobes". In an
embodiment, RGB Values assigned for the Ventricles are: R-233,
G-192, B-250. The server further enables the user to select
"Contour Segmentation mode" i.e., Semi-automatic segmentation. The
contour segmentation allows the user to select semi-automatic
active contour segmentation and start the semi-automatic
segmentation as shown in FIG. 26a. The contour segmentation enables
the user to adjust the boundaries of the region of interest
covering the Occipital Lobes. Once the `active label` is assigned
as "Occipital Lobes", one or more first images comprising the
Occipital Lobes are rendered in at least one anatomical plane and a
three-dimensional format. The server enables the user to identify
the Occipital Lobes in the at least one anatomical plane.
Once the Occipital Lobes is identified, the server enables the user
to mark a region of interest covering the Occipital Lobes and check
whether the Occipital Lobes is covered in the anatomical planes as
shown in FIG. 26b. The server provides a "segment 3D" tool that
allows the user to start semi-automatic segmentation. The user upon
clicking the "segment 3D" tool a toolbar appears. The toolbar
provides a "pre-segmentation" tool. The "pre-segmentation" tool
provides a "thresholding", "classification", "clustering", and
"edge attraction". The user can select any of the four tools. For
example, the "classification" tool is selected by the user.
Before starting the "classification" type segmentation, the server
enables the user to choose a label indicating "Occipital Lobes"
under the "Segmentation labels" tool. The "active label" tool under
"Segmentation label" tool is used to record and save information
(e.g., volumes, boundaries, manual edits performed to the
segmentation, etc.) obtained as a result of performing the
segmentation performed on the at least one structure i.e.,
Occipital Lobes in this case. The server further provides a "brush"
tool that allows the user to select appropriate brush (e.g., round
brush) and appropriate brush size to mark the brain structures
(e.g., GM, WM, frontal cortex) using the "Occipital Lobes" label of
the "Segmentation labels" as shown in FIG. 26c. The Brain structure
marked comprises brain structures (e.g., GM, WM, frontal cortex).
The "classification" tool allows the user to classify between
Occipital lobes and Intracranial Volume (ICV) by providing two
labels "Occipital Lobes" and "ICV". The "Occipital lobes" label is
used to classify between white matter and grey matter. The "ICV"
label is used to classify between dura, skull bone, ventricles or
csf. In an embodiment, if there is an error, the "classification"
tool further allows the user to add a third label as "Ventricles"
to classify the Ventricles separately. The different tissue samples
comprise white matter (WM) and grey matter (GM).
The server allows the user to mark tissue samples such as the WM
and GM. The server further provides a "train classifier" tool that
allows the user to train the classifier by clicking on the "train
classifier" tool. The server further renders a speed image that
shows the classification. The "train classifier" assigns a
probability value to a voxel belonging to the "foreground" class
vs. belonging to all other classes. Once the classifier is trained
using the manual segmentation (i.e., marking and differentiating
the tissue samples, Occipital Lobes, ICV, and the Ventricles) the
classifier automatically segments at least one structure within the
one or more first images of a different patient in future based on
micro-ethnicity information, age and gender. The server provides a
"Next" tab to complete the process.
The server provides a "Add bubble at cursor" tool as shown in FIG.
26d that allows the user to populate bubbles of appropriate sizes
exactly in the Occipital Lobes in at least three anatomical planes
to exactly extract the volume of the Occipital Lobes as shown in
FIG. 26e. Further the server provides a "Bubble radius" slider that
allows the user to vary the size of the bubbles. The server further
provides an "active bubbles" drop down menu that shows the bubbles
and its radius that are active. The server allows the user to add a
sufficient number of bubbles in the Occipital Lobes. The server
provides a "Next" tab to finalize bubbles' evolution.
The server provides a "continuous update" in a three-dimensional
window that enables to continuously update contour evolution. The
server further provides a "play" tab that allows the user to run,
play, and pause Active Contour Evolution. The server further
provides a "finish" tab that allows the user to submit when the
active contour evolution is done. In an embodiment, the server
allows the user to change the "active label" to "clear label" and
delete the voxels when the active contour evolution goes out of the
boundaries of the Occipital Lobes. The server allows the user to
edit the voxels by accessing the "brush" tool and selecting
appropriate brush and appropriate brush size. In another
embodiment, the server allows the user to change the "active label"
to "Occipital Lobes" and add the voxels when the active contour
evolution has not reached any part of the Occipital Lobes. The
server allows the user to edit the voxels by accessing the "brush"
tool and selecting appropriate brush and appropriate brush
size.
The server may render and save the Right Occipital Lobes in at
least one anatomical pane and in three-dimensional format under the
"Active Label" as "Occipital Lobes". Once the segmentation and
volume extraction of the Occipital Lobes are complete, the server
enables the user to save the one or more first images, the
workspace, the mesh and the one or more segmented image with a
patient id name. The server enables the user to place the directory
location for the image files in the patient ID folder. The server
enables the user to capture screenshots of the segmented image at
all anatomical planes in such a way that Occipital Lobes are
displayed well with labels. The server further allows the user to
capture screenshots of the segmented image in a three-dimensional
image format with the Occipital Lobes zoomed in and displayed
well.
The server renders the Occipital lobes in a three-dimensional
format. The server depicts the Occipital lobes with an arrow
pointing at one of the left side and right side as shown in FIG.
26f. The server provides a cursor and places it on a longitudinal
fissure. The server provides a "split" tool that enables the user
to place a line which traces the longitudinal fissure. The server
further renders an arrow on the three-dimensional window. The
server allows the user to select the "active label" as "Left
Occipital Lobes" when the arrow is pointing at the right side as
shown in FIG. 26g. The server allows the user to select the "active
label" as "Right Occipital Lobes" when the arrow is pointing at the
left side. The RGB values for the Left Occipital Lobes are: R-169,
G-176, B-136. Similarly, the server allows the user to select the
"active label" as "right Occipital Lobes" when the arrow is
pointing at the left side. The server provides an "accept" tab that
allows the user to accept and update the segmentation of 3D view
format. The server further enables the user to Check in the Axial
plane whether the Left Occipital Lobes has been labelled
properly.
FIG. 26h are reference figures provided for the segmentation that
illustrates the following. The lobes rest on the Tentorium
Cerebelli, a process of dura mater that separates the Cerebrum from
a Cerebellum. The lobes are structurally isolated in their
respective Cerebral hemispheres by the separation of the Cerebral
Fissure. The Parieto-Occipital Sulcus separates the Parietal and
Occipital Lobes. The lateral side is differentiated by the Lateral
Parietotemporal line.
FIG. 27a-27g illustrate a process of segmentation of Temporal
Lobes, according to one or more embodiments. The process of
segmentation of the Temporal Lobes comprises the following
technical steps. While performing the segmentation, a server
enables a user to select "active label" as "Temporal Lobes". The
server further enables the user to select "Contour Segmentation
mode" i.e., Semi-automatic segmentation. The contour segmentation
allows the user to select semi-automatic active contour
segmentation and start the semi-automatic segmentation as shown in
FIG. 27a. The contour segmentation tool enables a user to adjust
the boundaries of a region of interest covering the Temporal Lobes.
Once the `active label` is assigned as "Temporal Lobes", one or
more first images are rendered in at least one anatomical plane and
a three-dimensional format. The server enables the user to identify
the Temporal Lobes in the at least one anatomical plane.
Once the Temporal Lobes is identified, the server enables the user
to mark a region of interest covering the Temporal Lobes and check
whether the Temporal Lobes is covered in the anatomical planes as
shown in FIG. 27b. The server provides a "segment 3D" tool that
allows the user to start semi-automatic segmentation. The user upon
clicking the "segment 3D" tool a toolbar appears. The toolbar
provides a "pre-segmentation" tool. The "pre-segmentation" tool
provides a "thresholding", "classification", "clustering", and
"edge attraction". The user can select any of the four tools. For
example, the "classification" tool is selected by the user.
Before starting the "classification" type segmentation, the server
enables the user to choose a label indicating "Right Temporal
Lobes" under the "Segmentation labels" tool. The "Segmentation
label" tool is used to record and save information (e.g., volumes,
boundaries, manual edits performed to the segmentation, etc.)
obtained as a result of the segmentation performed on the at least
one structure i.e., Temporal Lobes in this case. The RGB values for
the right temporal lobe are: R-102, G-205, B-130.
The server further provides a "brush" tool that allows the user to
select appropriate brush (e.g., round brush) and appropriate brush
size to mark the brain structures (e.g., GM, WM, frontal cortex)
using the "Right Temporal Lobes" label of the "Segmentation
labels". The Brain structure marked comprises brain structures
(e.g., GM, WM, frontal cortex). The "classification" tool allows
the user to classify between Right Temporal lobes and Intracranial
Volume (ICV) by providing two labels "Right Temporal Lobes" and
"ICV". The "Right Temporal lobes" label is used to classify between
white matter and grey matter. The "ICV" label is used to classify
between dura, skull bone, ventricles or csf. In an embodiment, if
there is an error, the "classification" tool further allows the
user to add a third label as "Ventricles" to classify the
Ventricles separately. The different tissue samples comprise white
matter (WM) and grey matter (GM).
The server allows the user to mark tissue samples such as the WM
and GM as shown in FIG. 27c. The server further provides a "train
classifier" tool that allows the user to train the classifier by
clicking on the "train classifier" tool. The server further renders
a speed image that shows the classification. The "train classifier"
tool assigns a probability value to a voxel belonging to the
"foreground" class vs. belonging to all other classes. Once the
classifier is trained using the manual segmentation (i.e., marking
and differentiating the tissue samples, Right Temporal Lobes, ICV,
and the Ventricles) the classifier automatically segments at least
one structure within the one or more first images of a different
patient in future based on micro-ethnicity information, age and
gender and the like. The server provides a "Next" tab to complete
the process.
The server provides a "Add bubble at cursor" tool that allows the
user to populate bubbles of appropriate sizes exactly in the Right
Temporal Lobes, as shown in FIG. 27d, in at least three anatomical
planes to exactly extract the volume of the Right Temporal Lobes.
Further the server provides a "Bubble radius" slider that allows
the user to vary the size of the bubbles. The server further
provides an "active bubbles" drop down menu that shows the bubbles
and its radius that are active. The server allows the user to add a
sufficient number of bubbles in the Right Temporal Lobes as shown
in FIG. 27e. The server provides a "Next" tab to finalize bubbles'
evolution.
The server provides a "continuous update" that enables it to
continuously update contour evolution. The server further provides
a "play" tab that allows the user to run, play, and pause Active
Contour Evolution as shown in FIG. 27f. The server further provides
a "finish" tab that allows the user to submit when the active
contour evolution is done. In an embodiment, the server allows the
user to change the "active label" to "clear label" and delete the
voxels when the active contour evolution goes out of the boundaries
of the Right Temporal Lobes. The server allows the user to edit the
voxels by accessing the "brush" tool and selecting appropriate
brush and appropriate brush size. In another embodiment, the server
allows the user to change the "active label" to "Right Temporal
Lobes" and add the voxels when the active contour evolution has not
reached any part of the Right Temporal Lobes. The server allows the
user to edit the voxels by accessing the "brush" tool and selecting
appropriate brush and appropriate brush size.
The server may render and save the Right Temporal Lobes in at least
one anatomical pane and in three-dimensional format under the
"Active Label" as "Right Temporal Lobes". Once the segmentation and
volume extraction of the Right Temporal Lobes are complete, the
server enables a user to save one or more first images, the
workspace, the mesh and the one or more segmented image with a
patient id name. The server enables a user to place the directory
location for the image files in the patient ID folder. The server
enables the user to capture screenshots of the segmented image at
all anatomical planes in such a way that Right Temporal Lobes are
displayed well with labels. The server further allows the user to
capture screenshots of the segmented image in a three-dimensional
image format with the Right Temporal Lobes zoomed in and displayed
well. The server is configured to repeat the above steps for the
Left Temporal Lobes. The RGB values for the Left Temporal Lobes
are: R-210, G-140, B-206. The server provides an "accept" tab that
allows the user to accept and update the segmentation of 3D view
format. FIG. 27g illustrates the boundaries of the Temporal Lobes.
The lobe extends superiorly to Sylvian fissure. Posteriorly, the
lobe is differentiated by Lateral Parietotemporal line, which
separates the Temporal Lobe from Inferior Parietal Lobule of the
Parietal Lobe superiorly and the Occipital Lobe inferiorly.
FIGS. 28a and 28b illustrate a structure-based analysis report
showing a structure-based analysis, according to one or more
embodiments. The structure-based analysis report comprises a
patient details section, a snippet section, a volumetric analysis
section, a feature, and a volumetric derived analysis section. The
feature comprises at least one of the one or more volumes of the
region of interest (ROI), a cortical thickness, an atrophy
percentage, an asymmetry index score, a subfield volumetry of the
region of interest, annular volume changes, a progressive
supranuclear palsy (psp) index score, a magnetic resonance
perfusion imaging (MRPI) score, a frontal horn width to
intercaudate distance ratio (FH/CC), a medial temporal lobe atrophy
(MTA) score, a global cortical atrophy (GCA) scale, identification
of Intracranial bleeds, hemorrhage, microbleeds and their volume
analysis, a fracture detection, a midline shift identification, a
measurement of the midline shift identification and the at least
one structure with respect to the midline shift identification,
identifying a pathology associated with the at least one structure,
classifying the pathology identified, a tissue density
identification, an infarct identification, a Penumbra-core-viable
tissue identification, classification and volume calculation,
diffusion-weighted imaging (DWI) maps and apparent diffusion
coefficient (ADC) maps of the at least one structure, perfusion
maps comprising resting state functional magnetic resonance imaging
(rsfMRI), an alberta stroke programme early CT score (ASPECTS)
calculation, a collateral detection, a mismatch ratio calculation,
an angiography labeling and/or annotation, a large vessel occlusion
(LVO) detection, an Hypoperfusion index calculation, Diffusion
tensor imaging (DTI) fiber tracks, neural pathway connectivity
maps, correlation between a signal input, an image input and the
text input, classifying the signal input, identifying a normal
signal, identifying an abnormal signal, identifying a pre-ictal
signal, identifying an ictal signal, extracting symptoms, and
grading of condition specific effects. The patient details section
comprises information related to a patient such as an age, a
gender, a site ID, a patient ID, a patient name, a patient contact,
an exam date, and referring physician information (e.g., a
referring physician ID, referring physician name, etc.). The
snippet section may comprise quantitative volume such as whole
brain volume, an intracranial volume (ICV), and a brief summary of
output of the analysis report. The brief summary may comprise
clinical information indicating volume loss and volume loss
percentage, if any. The clinical information comprises symptoms,
existing conditions, etc. The brief summary further indicates
abnormalities in volumes of at least one structure. The brief
summary may also indicate normality in volumes of at least one
structure.
The volumetric analysis section shows volumes of at least one
structure, volume as % ICV, and reference range of the at least one
structure. The reference range is estimated between 25th and 95th
percentile. The volumetric derived analysis section shows one or
more derived analyzes that are derived from the one or more volumes
extracted. The volumetric derived analysis shows the corresponding
output and reference values. The volumetric derived analysis
section may comprise age expected atrophy, total atrophy
percentage, Whole Brain, Hippocampus. Hippocampus Asymmetry index,
Lobar Asymmetry index, annual volume changes, etc.
The structure-based analysis report may also comprise a graphical
representation of volumetric changes. The graphic representation of
volumetric changes depicts a time series volumetric changes over
time that aids physicians in their assessment of a patient's
prognosis and diagnosis. The analysis report may also comprise a
pictorial representation of volumes. The pictorial representation
of volumes shows orientation, position, shape, and volumes of at
least one structure within the segmented image. The pictorial
representation of volumes depicts the segmented image in at least
one of a three-dimensional format, and an anatomical plane.
FIG. 29a-29c illustrate an integrated analysis report showing an
integrated analysis of an image input, a text input, and a signal
input, according to one or more embodiments. The integrated
analysis report depicts a cognitive test result prepared by
performing the integrated analysis of the image input, the text
input, and the signal input.
The integrated analysis report shows the one or more physiological
signals comprising at least one spike that indicates if there are
any abnormalities. The integrated analysis report may show the
feature, and the one or more volumes in a three-dimensional format,
and at least one anatomical plane that is rendered to the user to
investigate, analyze, edit, and/or correct the one or more volumes.
The integrated analysis report renders the one or more
physiological signals comprising at least one spike and the one or
more volumes, and the one or more quantitative volumes parallelly.
This aids the user (e.g., physician) to perform predictive
prognosis, diagnosis and predict atrophy changes. The integrated
analysis report further aids the user to ensure that an analysis,
determined via a first input (e.g., the image input), is correct by
performing the analysis determined via a second input (e.g., the
signal input). In other words, the analysis report enables the user
to ensure the accuracy of the output by comparing the outputs of
the patient in a first dimension (e.g., obtained via the image
input), and a second dimension (e.g., obtained via the signal
input).
The integrated analysis report comprises at least one of the
feature, a patient details section, a snippet section, a volumetric
analysis section, and a volumetric derived analysis section. The
patient details section comprises information related to a patient
such as an age, a gender, a site ID, a patient ID, a patient name,
a patient contact, an exam date, and referring physician
information (e.g., a referring physician ID, referring physician
name, etc.). The snippet section may comprise clinical information.
The clinical information comprises symptoms, existing conditions,
additional info (applicable history/family history, etc.), and
cognitive function test, etc. The integrated analysis report
further renders cognitive test output. The analysis report further
comprises the Mill volumetric analysis section. The MRI volumetric
analysis section comprises pictorial representation of one or more
volumes in at least one anatomical plane. The MRI volumetric
analysis section further renders cortical image and at least one
segmented image.
The integrated analysis report further comprises a cortical
analysis section, and structural volumetric analysis section. The
cortical analysis section comprises absolute volumes and relative
volumes of the at least one structure such as cerebrospinal fluid,
grey matter, white matter, and cortical thickness. The structural
volumetric analysis section comprises quantitative volume (ml),
volume as ICV % and reference range for at least one structure. The
reference range may be the 25th percentile and the 95th percentile.
The structural volumetric analysis section further comprises
graphical representation illustrating the recorded volume with
respect to age. The graphical representation clearly illustrates a
plot of the recorded volume, the 25th percentile, the 95th
percentile and the 50th percentile that aids the doctors in their
assessment.
The integrated analysis report further comprises a volumetric
derived analysis section. The volumetric derived analysis section
indicates analysis, output and reference range. The integrated
analysis report further comprises a time series volumetric changes
section. The time series volumetric changes section depicts
graphical representation that shows changes in volumetric
measurement of the at least one structure over time. The time
series volumetric changes section aids the physicians in performing
their assessment. The integrated analysis report further comprises
a Diffusion Tensor Imaging (DTI) and functional output section that
indicates at least one of normality or abnormality in at least one
organ. The DTI and functional output section comprise at least one
structure in three-dimensional format and at least one anatomical
plane. The DTI and functional output section indicate functional
mapping of the at least one organ based on an imaging technique
(e.g., Blood oxygenation level dependent (BOLD) imaging technique)
to map different connectivity maps. The DTI and functional output
section help in understanding disease affected areas and related
cognitive/functional deficits for different functionalities such as
executive and summary, motivation, decision-making, attention,
orientation, etc.
The integrated analysis report further comprises a signal
monitoring and recording section. The signal monitoring and
recording section records at least one region of the one or more
physiological signals that comprises spike indicating one of
abnormality or normality in functioning of the at least one organ.
The signal monitoring and recording section highlights and
quantifies the region comprising at least one spike (for alpha,
Beta and gamma waves) that is responsible for the abnormality or
normality of the at least one organ with correlated temporal
resolution. The analysis report further comprises a summary of
findings. The summary of findings shows symptoms, at least one
spike information, etc. from the at least one spike.
FIGS. 30a-30b illustrate an EEG detailed report, according to one
or more embodiments. The EEG detailed report comprises a feature,
an information/recording conditions section, a
modulators/procedures section, a findings section, polygraphy
channels section, a summary of findings section and a screenshot
section. The information/recording conditions section comprises
information pertinent to recording time, recording date, recording
period, study ID, medical condition, alertness, etc. The
modulators/procedures section comprises sensitivity information
such as eye-closure sensitivity and other responses acquired from
the patient. The modulators/procedures section may also comprise
event related potential responses received from a patient. The
findings section comprises background activity, sleep/drowsiness,
Interictal findings, episodes, artifacts, if any. The background
activity comprises information related to background such as
amplitude, frequency, activity in response to the signal, etc. The
sleep/drowsiness comprises sleep stages. The artifacts section
comprises artifacts information such as low voltage, and lack of
compliance. The polygraphy channels section comprise channel
information acquired through one or more channels. The summary of
findings section comprises symptoms, medical conditions, regions
having at least one spike indicating abnormality, etc. and
conclusion of the findings. The summary of findings section further
comprises diagnostic significance, and clinical components. The
clinical components section describes the diagnosis that is
suitable for this patient based on the inference done. The
screenshot section comprises screenshots of at least one
physiological signal supporting the above findings and other
inferences illustrated in the EEG detailed analysis report. The EEG
detailed report further comprises graphical representation of EEG
reports along with ICU monitoring of at least one spike and/or pre
ictal detection. The EEG detailed report further comprises
information comprising spectral analysis.
FIG. 31 illustrates monitoring of one or more physiological
signals, according to one or more embodiments. A server is capable
of receiving one or more physiological signals in real-time. The
server may also receive the one or more physiological signals that
are acquired and recorded previously. The server analyzes and
monitors the one or more physiological signals for a predefined
period of time. The server is pre-trained with the one or more
physiological signals without any abnormal spikes and the one or
more physiological signals having abnormal spikes. The server then
analyzes the one or more physiological signals and detects at least
one spike and/or pre-ictal that indicates abnormality, if any. The
server upon finding the at least one spike, indicating abnormality,
highlights a region, comprising the at least one spike using one or
more identifiers with correlated temporal resolution, to the user
to aid in their assessment. The server is also capable of
performing spectral analysis and indicating amplitude, frequency
and other related parameters that may impact the functioning of the
at least one organ.
FIG. 32a illustrates a screenshot of a user interface that allows a
user to upload patient details, according to one or more
embodiments. A server, via a user interface depicted herein, allows
the user to provide the inputs such as at least one of an image
input, a text input, and a signal input. The server via the user
interface allows the user to drag and drop and/or select one or
more first images from an image source. The server, via the user
interface, allows the user to enter the text input such as name,
age, gender and select symptoms, micro-ethnicity, and medical
conditions from a drop-down box. The server also allows the user to
enter a referring physician's name. The server further provides a
`submit` tab that allows the user to submit the inputs and create a
record once the inputs are provided.
FIG. 32b illustrates a screenshot of a user interface that allows a
user to view patient details, according to one or more embodiments.
The user herein may be a radiologist. The user interface view,
depicted herein, provides a `View patients` tab that allows the
user to view patient details. The user interface shows a site ID, a
patient age, a patient gender, a symptom, a medical condition, and
an exam date when the user has accessed the `View patients` tab.
The site ID comprises information that helps to recognize a site
from where the inputs are scanned, obtained and/or uploaded. The
exam date refers to a date at which the patient is examined. The
user interface depicts the patient details. However, identification
information (e.g., facial information, name, locality, address,
etc.) that are adapted to recognize an identity of the patient is
anonymized i.e., not shown to the user.
The user interface further shows volumetric analysis of at least
one structure or organ (e.g., cardiovascular organ, neural organ,
orthopedic organ, etc.) of the patient. The volumetric analysis
shows one or more quantitative volumes associated with the at least
one structure or the organ. For example, when the inputs (e.g., the
text input, the image input, and the signal input) related to brain
is uploaded, the volumetric analysis shows the one or more
quantitative volumes of the at least one structure associated with
the brain such as intracranial volume (ICV), whole brain,
ventricles, Lt. Hippocampus, Rt. Hippocampus, Lt. Temporal Lobe,
Rt. Temporal Lobe, etc. The user interface further depicts the one
or more quantitative volumes such as volumes in ml, volume as % ICV
for the at least one structure and reference ranges for the at
least one structure. The user interface also highlights the one or
more quantitative volumes with a different color to readily
identify and indicate that the values are out of the reference
range.
FIG. 32c illustrates a screenshot of a user interface rendering a
segmented image, according to one or more embodiments. The user
interface shown in FIG. 32d renders the segmented image in a
three-dimensional format and an anatomical plane. The server, via
the user interface, allows the user to select and view a portion of
the segmented image in an enhanced view. The enhanced view may be a
zoomed view. The anatomical plane may comprise at least one of a
parasagittal plane, a sagittal plane, a coronal plane, and an axial
plane, etc. The server further allows the user to select the
different anatomical plane and render the segmented image in the
different anatomical plane. The server further allows the user to
readily identify orientation, position, shape, and volumes of at
least one structure within the segmented image and other
information such as age, gender, ICV, and other micro-ethnicity
information that may impact the volumes of the at least one
structure.
FIG. 32d illustrates a screenshot of a user interface that allows a
user to view patient details, according to one or more embodiments.
The user herein may be a manager. The user interface, depicted
herein, provides a `View patients` tab that allows the user to view
patient details. The user interface shows a site ID, a patient age,
a patient gender, a symptom, a medical condition, and an exam date
when the user has clicked the `View patients` tab. The site ID
comprises information to recognize a site from where the inputs are
scanned, obtained and/or uploaded. The exam date refers to a date
at which the patient is examined. The user interface renders the
patient details. However, identification information (e.g., facial
information, name, locality, address, etc.) that are adapted to
recognize an identity of the patient, either digitally or manually,
is anonymized i.e., not shown to the user.
The user interface further shows volumetric analysis of at least
one structure or organ (e.g., cardiovascular organ, neural organ,
orthopedic organ, etc.) of the patient. The volumetric analysis
shows one or more quantitative volumes associated with the at least
one structure or the organ. For example, when the inputs (e.g., the
text input, the image input, and the signal input) related to brain
is uploaded, the volumetric analysis shows the one or more
quantitative volumes of the at least one structure associated with
the brain such as intracranial volume (ICV), whole brain,
ventricles, Lt. Hippocampus, Rt. Hippocampus, etc. The user
interface further depicts the one or more quantitative volumes such
as quantitative volumes (ml), volume as % ICV for the at least one
structure, and reference ranges for the at least one structure. The
user interface also highlights the one or more quantitative volumes
with a different color to readily identify and indicate that the
values are out of the reference range.
FIG. 33 illustrates processing of EEG signals, according to one or
more embodiments. A server processing the EEG signal input
comprises a) accepting data, b) pre-processing the data, c) data
representation, d) post processing the data, e) EEG MRI overlay, f)
report generation, g) cloud storage, h) building ERP pipeline for
Dementia and neurodegeneration, and i) ICU monitoring. The server
accepts the EEG signal from hardware itself where the server
integrates with the hardware itself and accepts raw data in the
form of EDF. In another embodiment, the EEG signal will be pushed
to the server by the technician/doctor via a web application. The
pre-processing of the raw data comprises sequential steps such as
importing raw data, event markers, artifacts removal,
interpolation, channel type selection/channel location,
referencing/re-referencing, filtering, epoching, characteristics
extraction and characteristics selection,
classification/statistical manipulation and result evaluation. The
data once pre-processed can be represented in several ways both in
frequency domain and time domain. The most common format of data
representation is in the form of a sinusoidal graph. The
representation of the EEG signal in various formats assists in
deeper understanding of the patient's condition. The formats
comprise spectral analysis graphs, Fast Fourier Transform (FFT)
analysis, amplitude graphs, asymmetry graphs, spatial
representation of EEG data, etc.
The post processing comprises characteristic Extraction/Selection,
that is to Identify statistically significant characteristics
(e.g.--Spike in epilepsy) using at least one of Multivariate time
series analysis, Wavelet transform (Time-frequency), Fourier
transform (Frequency Domain), Principal component analysis (Time
Domain), independent component analysis (ICA), etc. The post
processing is continued by optimal parameter and characteristics
set identification (e.g.,--characteristic shuffle analysis, ranking
characteristics) and classification/statistical Manipulation of the
signals applying various machine learning algorithms [e.g.,--Linear
discriminant analysis (LDA), Multi-layer perceptron (MLP), Support
vector machine (SVM), etc.]
The server further allows a user to overlay the EEG signal on the
MRI Image as a Heat map for better visualization of the signal. The
overlaying enables the user to better do surgical planning, because
they better understand where the seizure is originating, and it
helps them kind of understand the source of the seizure. The server
stores all patient related data in the cloud and enables the user
to access the data using a user ID. The server acquired event
related potentials (ERP) from the patient and builds clinical
endpoints in dementia and Mild cognitive impairment. The server
then generates a report once the clinician has entered the patient
history and has selected the at least one analysis. The server is
further capable of performing intensive care unit (ICU) monitoring
and detecting abnormalities. The server establishes a baseline of
the normal EEG signal of the patient and points out any
abnormalities in real time to alert the clinicians of any ongoing
seizures.
FIG. 34 illustrates a data flow of a system, according to one or
more embodiments. The data flow comprises sequential flow of data
as described below. At step 3402, an upload page is rendered to a
user by a server that allows the user to upload inputs upon logging
into the server. At step 3404, the server allows the user to
provide text inputs such as age, gender, symptom, medical
condition, referring physicians, micro-ethnicity, etc. and DICOM
format files. At step 3406, the server allows the user to submit
DICOM files upon providing the details. The server further checks
whether the DICOM format files have more than 1.5 Tesla and enables
the user to check whether DICOM format files have predefined
quality and quantity, at step 3408. The server also checks whether
the DICOM format files are 3D gradient echo (GRE) ISO sequences, at
step 3410. At step 3412, when the server detects upload error
(i.e., not having predefined quality and quantity) or incorrect
files, the server directs the user to the upload page again.
At step 3414, the server directs the uploaded DICOM format files to
pre-processing, when the server detects that the uploaded DICOM
files are optimum or good. At step 3416, the server converts the
DICOM format files to NIfTI format files for data anonymization. At
step 3418, the server performs denoising i.e., filtering noises
from the inputs. At step 3420 and 3422, the server performs bias
correction and matrix adjustments (reorient the image) respectively
to make the uploaded files suitable for image segmentation. At step
3424, the server stores the files in S3 bucket. At step 3426, the
server calls the uploaded files to the core application programming
interface (API) for segmentation. At step 3428, the server performs
image segmentation on the NIfTI format files. At step 3430, the
server records the segmented files under a worklist of the
corresponding user (e.g., physician, manager, admin etc.) as per
privileges granted by the server. At step 3432, the server is
enabled to extract image files from the S3 bucket and renders the
image files for viewing, analysis and editing purposes. At step
3434, the server displays the image files (e.g., segmented images,
analysis report, etc.) using a viewer configured (e.g., papaya
viewer).
FIG. 35 illustrates a workflow diagram of a server, according to
one or more embodiments. The workflow illustrates a sequential flow
performed by the server. At step 3502, the server renders an upload
page to a user upon login. At step 3504, the server receives the
uploaded inputs and records case details online in a database to be
accessed by the user. At step 3506, the server reorients the image.
At step 3508, the server defaces the images to recognize the
identity of the patient. At step 3510, the server then stores the
inputs in S3 bucket for retrieval, editing, viewing, and future
uses. At step 3512, the server calls core application programming
interface (API) to process the case. At step 3514, the server
downloads the case details from the S3 bucket. At step 3516, the
server performs image segmentation. At step 3518, the server
uploads the segmented images to S3 bucket. The uploaded segmented
images may be utilized for study, analysis, investigation,
volumetric extraction, volumetric analysis, atrophy, and predictive
prognosis and diagnosis. In an embodiment, the server performs
multimodal analysis and cross checks with other modes of analysis
and ensures the accuracy of predictive prognosis, diagnosis and
atrophy determination.
FIG. 36 further illustrates an architecture of a system, according
to one or more embodiments. The architecture depicts that a user
can communicate with a server through one of a product interface
3602, and a web interface 3604. The server may comprise a platform
3606 and core API 3608. The core API 3608 comprises segmentation
API 3610 and core algorithm 3612. The core algorithm 3612 is
configured to handle requests, coordinate functions, etc. The
segmentation API 3610 is configured to perform image segmentation
and other volumetric derived analysis. The platform 3606 comprises
an upload API 3614, persistence script 3616, reorientation script
3618, and defacer script 3620. The upload API 3614 is configured to
enable the user to upload the inputs such as image, text, and
signal inputs. The persistence script 3616 enables the server to
withstand optimum quality of the inputs. The reorientation script
3618 enables the server to reorient the images. The defacer script
3620 further enables the user to perform anonymization to break the
link between data and a given participant so that the participant
cannot be identified, directly or indirectly.
FIG. 37 illustrates an architecture of a system, according to one
or more embodiments. The architecture shown comprises one or more
computing units 3702 (A-N) communicating to a server via a
communication network. The communication network may be a wireless
communication network or a wired communication network. The
computing unit communicates to the server using a product interface
or a web interface though a secured internet gateway 3704. The
server comprises a public subnet 3706 and private subnet 3708. The
server, in an embodiment, comprises a graphical processing unit.
The public subnet 3706 and the private subnet 3708 are secured. The
private subnet 3708 comprises a core API 3710 and a platform API
3712. The platform API 3712 and the core API 3710 are already
described in FIG. 36. The one or more computing units 3702 A-N
communicate to the private subnets 3708 via the load balancers 3714
and 3716. The server stores the processed and raw inputs in an S3
bucket 3720. The public subnet may comprise a virtual private
network (VPN) server 3718.
FIG. 38a-38e illustrate an analysis report generated based on
condition specific analysis, according to one or more embodiments.
A server receives input as at least one of an image input, a text
input, and a signal input. Depending on the available data, the
server performs the multimodal analysis. The server may extract the
text input of relevant information using a natural language
processing module from a text or documents stored on the Hospital
Information System (HIS). The NLP module may extract relevant
information such as symptoms, clinical history (e.g., vitamin
deficiency, family history, genetic history, trauma, etc), and
cognitive test analysis like Computerized Cognitive Testing in
Epilepsy (CCTE), Montreal Score, Cambridge Neuro-psychological Test
Automated Battery (CANTAB), Mini Mental State Examination (MMSE),
Mini-Cog, and the like.
The server may also receive the text input such as an
electroencephalogram (EEG) signal. The server upon receipt of the
EEG signal may monitor the signal input and detect abnormalities in
the EEG signal with correlated temporal resolution either with
Structural Magnetic Resonance Imaging (sMRI) or others. The server
can acquire the EEG signal in real-time and monitoring abnormality
of the EEG signal. The server is also capable of correlating the
detected abnormality with other image input (such as scan images)
to double-check/ensure the abnormality in the patient's health. The
server receives the image input as at least one of sMRI, fMRI, CT,
DTI, PET, etc.
The sMRI may be used to perform structural volumetric analysis
based on 3D MRI correlated with normative population (specific to
ethnicity) as well as condition specific population, cortical
thickness analysis, subfield analysis, etc. The fMRI may be used to
perform functional mapping of the brain based on `BOLD` imaging
technique to map different connectivity maps. The fMRI image input
helps in understanding disease affected areas and related
cognitive/functional deficits. The fMRI has poor temporal
resolution and involves complex processing to understand which
connectivity networks are involved & affected. The server can
provide both fused images with structural MRI as well as automated
connectivity maps where the problematic areas will be pointed out
for physician's review. Upon receiving the CT input, the server
provides structural as well as perfusion-based analysis of the CT
images to derive a first look into the disease pattern.
The server receives the DTI and performs White matter tracts
analysis. White matter tracts analysis has become the core of
surgical planning in many conditions. The server provides automated
DTI analysis highlighting the changes in the tracts. The server may
receive the PET and perform functional analysis based on contrast
uptake. As PET provides good spatial resolution with poor temporal
information, the server help physicians understand temporal
information by fusing PET with MRI and produce such overlays which
can be visualized easily
The server also provides a user interface to upload patient details
in which users can enter a specific medical condition of a patient
(e.g., epilepsy). The server upon receiving the medical condition
enables it to perform a condition specific analysis. The condition
specific analysis is performed by following steps. Consider the
patient is having the medical condition as Epilepsy. The server
then compares age, gender, ICV, micro-ethnicity information of the
patient with a condition specific population i.e., (a population of
individuals having the medical condition as epilepsy). In one
embodiment, the server compares the information of the patient with
the normative population (i.e., wide analysis). The server, in this
embodiment, predicts a prognosis and analyzes the deterioration or
improvement in volumetric changes, quantitative volume, abnormality
of the patient.
In another embodiment, the server compares the information of the
patient with a condition specific population (i.e., population of
the individuals having the same medical condition as epilepsy)
i.e., narrow down analysis. The server, in this embodiment,
performs a prognosis, accurate diagnosis. The server, by performing
a condition specific analysis, can perform a predictive prognosis
over time, accurate diagnosis and comprehensive management of
patient's health. The comprehensive management of the patient's
health is performed by performing a predictive prognosis over
time.
For instance, consider the server has predicted a first prognosis
for a condition specific analysis for a first point of time. The
first prognosis is predicted for the first point of time
considering the medication information (e.g., medication that the
patient has intake during the first point of time) of the patient
and other relevant information. The first prognosis may be
performed via a multimodal analysis. The server has also predicted
a second prognosis for a condition specific analysis for a second
point of time. The second prognosis is predicted for the second
point of time considering the medication information (e.g.,
medication that the patient has intake during the second point of
time) of the patient and other relevant information. The second
prognosis may be performed via a multimodal analysis. The server is
also capable of determining deterioration or improvement in at
least one volumetric changes and quantitative volumes by comparing
the first prognosis and the second prognosis. The server determines
the deterioration or the improvement, in terms of percentage,
between the first prognosis and the second prognosis. The server is
then trained with different values of the deterioration or the
improvement over time. The server is then capable of determining
the deterioration or improvement in the volumetric changes and
quantitative volumes for a third point of time (in future) based on
the training provided. The server determines the deterioration or
the improvement in quantitative values for the third point of time.
The quantitative values of the deterioration or the improvement in
the future enables and assists the physicians to treat/change the
medication regime for the patient accordingly.
FIG. 38a-38e depicts the analysis report generated based on
condition specific analysis. Once the medical condition (e.g.,
epilepsy) is specified, the server compares the information (age,
gender, ICV, micro-ethnicity) of the patient with the condition
specific population (i.e., individuals who are having epilepsy
symptoms). The server then derives the 25th and the 95th percentile
which are then used as the customized references in performing the
predictive prognosis, accurate diagnosis and comprehensive
management. The quantitative volumes, and the volumes of the
patient which fall between the values of the 25th and the 95th
percentile are considered to be healthy/normal. The quantitative
volumes, and the volumes of the patient which falls beyond/outside
the 25th and the 95th percentile are considered to be
unhealthy/abnormal.
The analysis report shown in FIGS. 38a-38e is similar to FIGS. 28
and 29. The analysis report, shown in this embodiment, illustrates
a condition specific integrated analysis of image, signal and text
inputs. The analysis report comprises an output section which
clearly describes a clinically analytical output obtained from an
integrated and condition specific analysis. The output section
points out an abnormality with respect to each input. The analysis
report comprises a clinical information section which provides
details about symptoms, existing conditions, and cognitive function
test. An MRI volumetric analysis section renders an image of the
region of interest which aids the physician to examine the volumes
of the region of interest. The analysis report also renders
segmented images.
The analysis report also comprises a cortical analysis section
which comprises volume information of at least one of CSF, grey
matter, white matter, and cortical thickness. The analysis report
further comprises a structural volumetric analysis section which
comprises volumes of the structures, volume as ICV %, and their
reference ranges (i.e., 25th and 95th percentile). The analysis
report further comprises a graph indicating condition specific
population comparison with the recorded volumes of the patient. The
analysis report further comprises a volumetric derived analysis
section which indicates the differences in recorded volumes and the
reference ranges. The volumetric derived analysis section also
shows annual volume changes based on the derived analysis. The
analysis report further shows a graph indicating time series
volumetric changes at different points of time. The analysis report
further renders a DTI and functional output which provides
structural connectivity and functional connectivity information.
The DTI and functional output also render connectivity mapping. The
analysis report further comprises an EEG analysis section which
indicates/highlights abnormal spikes. The abnormal spikes may be
used to correlate with other temporal resolution either with sMRI
or other inputs and perform an integrated analysis.
The foregoing disclosure provides illustration and description, but
is not intended to be exhaustive or to limit the implementations to
the precise form disclosed. Modifications and variations are
possible in light of the above disclosure or may be acquired from
practice of the implementations.
Those skilled in the art will appreciate that the invention may be
practiced in network computing environments with many types of
computer system configurations, including, personal computers,
desktop computers, laptop computers, message processors, hand-held
devices, multi-processor systems, microprocessor-based or
programmable consumer electronics, network PCs, minicomputers,
mainframe computers, mobile telephones, PDAs, pagers, routers,
switches, etc. The invention may also be practiced in distributed
system environments where local and remote computer systems, which
are linked (either by hardwired data links, wireless data links, or
by a combination of hardwired and wireless data links) through a
network, both perform tasks. In a distributed system environment,
program modules, units may be located in both local and remote
memory storage devices.
The present invention may be embodied in other specific forms
without departing from its spirit or characteristics. The described
embodiments are to be considered in all respects only as
illustrative and not restrictive. The scope of the invention is,
therefore, indicated by the appended claims rather than by the
foregoing description. All changes which come within the meaning
and range of equivalency of the claims are to be embraced within
their scope.
Although the present embodiments have been described with reference
to specific example embodiments, it will be evident that various
modifications and changes may be made to these embodiments without
departing from the broader spirit and scope of the various
embodiments. For example, the various devices, units and modules
described herein may be enabled and operated using hardware
circuitry (e.g., CMOS based logic circuitry), firmware, software or
any combination of hardware, firmware, and software (e.g., embodied
in a non-transitory machine-readable medium). For example, the
various electrical structures and methods may be embodied using
transistors, logic gates, and electrical circuits (e.g.,
application specific integrated circuitry (ASIC) and/or Digital
Signal Processor (DSP) circuitry).
In addition, it will be appreciated that the various operations,
processes, and methods disclosed herein may be embodied in a
non-transitory machine-readable medium and/or a system.
Accordingly, the specification and drawings are to be regarded in
an illustrative rather than a restrictive sense.
INCORPORATION BY REFERENCE
All patents, patent application publications, and non-patent
literature mentioned in the application are incorporated by
reference in their entirety. G.B.D. Disease and Injury Incidence
and Prevalence Collaborators, "Global, regional, and national
incidence, prevalence, and years lived with disability for 328
diseases and injuries for 195 countries, 1990-2016: a systematic
analysis for the global burden of disease study 2016," Lancet, vol.
390, pp. 1211-1259, 2017; World Health Organization, Epilepsy: A
Public Health Imperative, World Health Organization, Geneva,
Switzerland, 2019; M. Prince, A. Wimo, M. Guerchet, G. C. Ali, Y.
T. Wu, and M. Prina, "The global impact of dementia. An analysis of
prevalence, incidence, cost and trends," World Alzheimer Report
2015, Alzheimer's Disease International (ADI), London, U K, 2015;
H. A. Born, "Seizures in Alzheimer's disease," Neuroscience, vol.
286, pp. 251-263, 2015; W. A. Hauser, M. L. Morris, L. L. Heston,
and V. E. Anderson, "Seizures and myoclonus in patients with
Alzheimer's disease," Neurology, vol. 36, no. 9, p. 1226, 1986; N.
Scarmeas, L. S. Honig, H. Choi et al., "Seizures in Alzheimer's
disease: who, when, and how common?" Archives of Neurology, vol.
66, no. 8, pp. 992-997, 2009; Y. Holler and E. Trinka, "What do
temporal lobe epilepsy and progressive mild cognitive impairment
have in common?" Frontiers in Systems Neuroscience, vol. 8, 2014;
P. Fischer, S. Jungwirth, S. Zehetmayer et al., "Conversion from
subtypes of mild cognitive impairment to Alzheimer dementia,"
Neurology, vol. 68, no. 4, pp. 288-291, 2007; C. B. Dodrill,
"Neuropsychological effects of seizures," Epilepsy & Behavior,
vol. 5, pp. 21-24, 2004; Y. Holler and E. Trinka, "Is there a
relation between EEG slow waves and memory dysfunction in epilepsy?
a critical appraisal," Frontiers in Human Neuroscience, vol. 9,
2015; M. D. Holmes, C. B. Dodrill, R. J. Wilkus, L. M. Ojemann, and
G. A. Ojemann, "Is partial epilepsy progressive? ten-year follow-up
of EEG and neuropsychological changes in adults with partial
seizures," Epilepsia, vol. 39, no. 11, pp. 1189-1193, 1998; H.
Stefan and E. Pauli, "Progressive cognitive decline in epilepsy: an
indication of ongoing plasticity," Progress in Brain Research, vol.
135, pp. 409-417, 2002; G. L. Holmes, "What is more harmful,
seizures or epileptic EEG abnormalities? is there any clinical
data?" Epileptic Disorders, vol. 16, no. NS1, pp. 12-22, 2014; B.
Bondi, N. Philippi, O. Bousiges et al., "Do we know how to diagnose
epilepsy early in Alzheimer's disease?" Revue Neurologique, vol.
173, no. 6, pp. 374-380, 2017; K. A. Vossel, A. J. Beagle, G. D.
Rabinovici et al., "Seizures and epileptiform activity in the early
stages of Alzheimer disease," JAMA Neurology, vol. 70, no. 9, pp.
1158-1166, 2013; D. Friedman, L. S. Honig, and N. Scarmeas,
"Seizures and epilepsy in Alzheimer's disease," CNS Neuroscience
& Therapeutics, vol. 18, no. 4, pp. 285-294, 2012; A. Horvath,
A. Szu}cs, G. Bares, J. L. Noebels, and A. Kamondi, "Epileptic
seizures in Alzheimer disease," Alzheimer Disease & Associated
Disorders, vol. 30, no. 2, pp. 186-192, 2016; K. A. Vossel, K. G.
Ranasinghe, A. J. Beagle et al., "Incidence and impact of
subclinical epileptiform activity in Alzheimer's disease," Annals
of Neurology, vol. 80, no. 6, pp. 858-870, 2016; S. Shorvon and E.
Trinka, "Nonconvulsive status epilepticus and the postictal state,"
Epilepsy & Behavior, vol. 19, no. 2, pp. 172-175, 2010; E.
Trinka, G. Kramer, and K. Werhahn, "Vascular precursor
epilepsy--old wine in new skins?" Epilepsy & Behavior, vol. 48,
pp. 103-104, 2015; K. A. Vossel, M. C. Tartaglia, H. B. Nygaard, A.
Z. Zeman, and B. L. Miller, "Epileptic activity in Alzheimer's
disease: causes and clinical relevance," The Lancet Neurology, vol.
16, no. 4, pp. 311-322, 2017; A. Bakker, G. L. Krauss, M. S. Albert
et al., "Reduction of hippocampal hyperactivity improves cognition
in amnestic mild cognitive impairment," Neuron, vol. 74, no. 3, pp.
467-474, 2012; A. D. Lam, G. Deck, A. Goldman, E. N. Eskandar, J.
Noebels, and A. J. Cole, "Silent hippocampal seizures and spikes
identified by foramen ovale electrodes in Alzheimer's disease,"
Nature Medicine, vol. 23, no. 6, pp. 678-680, 2017; H. Hampel, S.
J. Teipel, and K. Burger, "Neurobiologische fruhdiagnostik der
alzheimer-krankheit," Der Nervenarzt, vol. 78, no. 11, pp.
1310-1318, 2007; P. Suppa, H. Hampel, T. Kepp et al., "Performance
of hippocampus volumetry with FSL-FIRST for prediction of
Alzheimer's disease dementia in at risk subjects with amnestic mild
cognitive impairment," Journal of Alzheimer's Disease, vol. 51, no.
3, pp. 867-873, 2016; A. C. Burggren, B. Renner, M. Jones et al.,
"Thickness in entorhinal and subicular cortex predicts episodic
memory decline in mild cognitive impairment," International Journal
of Alzheimer's Disease, vol. 2011, Article ID 956053, 9 pages,
2011; G. C. Chiang, P. S. Insel, D. Tosun et al., "Identifying
cognitively healthy elderly individuals with subsequent memory
decline by using automated MR temporoparietal volumes," Radiology,
vol. 259, no. 3, pp. 844-851, 2011; J. J. Gomar, M. T.
Bobes-Bascaran, C. Conejero-Goldberg, P. Davies, and T. E.
Goldberg, "Utility of combinations of biomarkers, cognitive
markers, and risk factors to predict conversion from mild cognitive
impairment to Alzheimer disease in patients in the Alzheimer's
disease neuroimaging initiative," Archives of General Psychiatry,
vol. 68, no. 9, pp. 961-969, 2011; M. M. Mielke, O. C. Okonkwo, K.
Oishi et al., "Fornix integrity and hippocampal volume predict
memory decline and progression to Alzheimer's disease," Alzheimer's
& Dementia, vol. 8, no. 2, pp. 105-113, 2012; P. Vemuri, H. J.
Wiste, S. D. Weigand et al., "MRI and CSF biomarkers in normal,
MCI, and AD subjects: predicting future clinical change,"
Neurology, vol. 73, no. 4, pp. 294-301, 2009; H. Wolf, M. Grunwald,
G. M. Ecke et al., "The prognosis of mild cognitive impairment in
the elderly," Alzheimer's Disease--From Basic Research to Clinical
Applications, vol. 54, pp. 31-50, 1998; J. L. Woodard, M.
Seidenberg, K. A. Nielson et al., "Pre-diction of cognitive decline
in healthy older adults using fMRI," Journal of Alzheimer's
Disease, vol. 21, no. 3, pp. 871-885, 2010; S. Kovacevic, M. S.
Rafii, and J. B. Brewer, "High-throughput, fully automated
volumetry for prediction of MMSE and CDR decline in mild cognitive
impairment," Alzheimer Disease & Associated Disorders, vol. 23,
no. 2, pp. 139-145, 2009; S. Alam, G.-R. Kwon, J.-I. Kim, and C.-S.
Park, "Twin SVM-based classification of Alzheimer's disease using
complex dual-tree wavelet principal coefficients and LDA," Journal
of Healthcare Engineering, vol. 2017, Article ID 8750506, 12 pages,
2017; A. Ayaz, M. Z. Ahmad, K. Khurshid, and A. M. Kamboh, "MRI
based automated diagnosis of alzheimer's: fusing 3D
wavelet-features with clinical data," in Proceedings of the 2017
39th Annual International Conference of the IEEE Engineering in
Medicine and Biology Society (EMBC), pp. 1210-12130, Seogwipo,
South Korea, July 2017; S. Wang, Y. Chen, Y. Zhang, E. Lee, Z.
Dong, and P. Phillips, "3D-DWT improves prediction of AD and MCI,"
in Proceedings of the First International Conference on Information
Science and Electronic Technology, ISET 2015, pp. 60-63, Wuhan,
China, March 2015; A. L. Dallora, S. Eivazzadeh, E. Mendes, J.
Berglund, and P. Anderberg, "Machine learning and microsimulation
techniques on the prognosis of dementia: a systematic literature
review," PLoS One, vol. 12, no. 6, Article ID e0179804, 2017; D. S.
Goodin, "Electrophysiologic evaluation of dementia," Neurologic
Clinics, vol. 3, no. 3, pp. 633-647, 1985; E.-L. Helkala, V.
Laulumaa, H. Soininen, J. Partanen, and P. J. Riekkinen, "Different
patterns of cognitive decline related to normal or deteriorating
EEG in a 3-year follow-up study of patients with Alzheimer's
disease," Neurology, vol. 41, no. 4, p. 528, 1991; H. Soininen, J.
Partanen, V. Laulumaa, E.-L. Helkala, M. Laakso, and P. J.
Riekkinen, "Longitudinal EEG spectral analysis in early stage of
Alzheimer's disease," Electroencephalography and Clinical
Neurophysiology, vol. 72, no. 4, pp. 290-297, 1989; V. Jelic, S.-E.
Johansson, O. Almkvist et al., "Quantitative electroencephalography
in mild cognitive impairment: longitudinal changes and possible
prediction of Alzheimer's disease," Neurobiology of Aging, vol. 21,
no. 4, pp. 533-540, 2000; A. Tsolaki, D. Kazis, I. Kompatsiaris, V.
Kosmidou, and M. Tsolaki, "Electroencephalogram and Alzheimer's
disease: clinical and research approaches," International Journal
of Alzheimer's Disease, vol. 2014, Article ID 349249, 10 pages,
2014; C. Babiloni, L. Benussi, G. Binetti et al., "Genotype
(cystatin c) and EEG phenotype in Alzheimer disease and mild
cognitive impairment: a multicentric study," NeuroImage, vol. 29,
no. 3, pp. 948-964, 2006; C. Babiloni, P. Bosco, R. Ghidoni et al.,
"Homocysteine and electroencephalographic rhythms in Alzheimer
disease: a multicentric study," Neuroscience, vol. 145, no. 3, pp.
942-954, 2007; K. Bennis, G. Rondouin, E. Benattar, A. Gabelle, and
J. Touchon, "Can event-related potential predict the progression of
mild cognitive impairment?" Journal of Clinical Neurophysiology,
vol. 28, no. 6, pp. 625-632, 2011; R. M. Chapman, J. W. McCrary, M.
N. Gardner et al., "Brain ERP components predict which individuals
progress to Alzheimer's disease and which do not," Neurobiology of
Aging, vol. 32, no. 10, pp. 1742-1755, 2011; C.-L. Lai, R.-T. Lin,
L.-M. Liou, and C.-K. Liu, "The role of event-related potentials in
cognitive decline in Alzheimer's disease," Clinical
Neurophysiology, vol. 121, no. 2, pp. 194-199, 2010; S. Jiang, C.
Qu, F. Wang et al., "Using event-related potential P300 as an
electrophysiological marker for differential diagnosis and to
predict the progression of mild cognitive impairment: a
meta-analysis," Neurological Sciences, vol. 36, no. 7, pp.
1105-1112, 2015; P. Missonnier, M.-P. Deiber, G. Gold et al.,
"Working memory load-related electroencephalographic parameters can
differentiate progressive from stable mild cognitive impairment,"
Neuroscience, vol. 150, no. 2, pp. 346-356, 2007; P. Missonnier, G.
Gold, L. Fazio-Costa et al., "Early event-related potential changes
during working memory activation predict rapid decline in mild
cognitive impairment," The Journals of Gerontology Series A:
Biological Sciences and Medical Sciences, vol. 60, no. 5, pp.
660-666, 2005; J. M. Olichney, J. R. Taylor, J. Gatherwright et
al., "Patients with MCI and N400 or P600 abnormalities are at very
high risk for conversion to dementia," Neurology, vol. 70, no. 19,
pp. 1763-1770, 2008; V. T. Papaliagkas, G. Anogianakis, M. N.
Tsolaki, G. Koliakos, and V. K. Kimiskidis, "Combination of P300
and CSF .beta.-amyloid (1-42) assays may provide a potential tool
in the early diagnosis of Alzheimer's disease," Current Alzheimer
Research, vol. 7, no. 4, pp. 295-299, 2010; S. Elmstahl and I.
Rosen, "Postural hypotension and EEG variables predict cognitive
decline: results from a 5-year follow-up of healthy elderly women,"
Dementia and Geriatric Cognitive Disorders, vol. 8, no. 3, pp.
180-187, 1997; A. A. Gouw, A. M. Alsema, B. M. Tijms et al., "EEG
spectral analysis as a putative early prognostic biomarker in
non-demented, amyloid positive subjects," Neurobiology of Aging,
vol. 57, pp. 133-142, 2017; C. Huang, L.-O. Wahlund, T. Dierks, P.
Julin, B. Winblad, and V. Jelic, "Discrimination of alzheimer's
disease and mild cognitive impairment by equivalent EEG sources: a
cross-sectional and longitudinal study," Clinical Neurophysiology,
vol. 111, no. 11, pp. 1961-1967, 2000; C. Luckhaus, B.
Grass-Kapanke, I. Blaeser et al., "Quantitative EEG in progressing
vs stable mild cognitive impair-ment (MCI): results of a 1-year
follow-up study," International Journal of Geriatric Psychiatry,
vol. 23, no. 11, pp. 1148-1155, 2008; P. Missonnier, G. Gold, F. R.
Herrmann et al., "Decreased theta event-related synchronization
during working memory activation is associated with progressive
mild cognitive impairment," Dementia and Geriatric Cognitive
Disorders, vol. 22, no. 3, pp. 250-259, 2006; F. Nobili, F.
Copello, P. Vitali et al., "Timing of disease progression by
quantitative EEG in Alzheimer's patients," Journal of Clinical
Neurophysiology, vol. 16, no. 6, pp. 566-573, 1999; S.-S. Poil, W.
de Haan, W. M. van der Flier, H. D. Mansvelder, P. Scheltens, and
K. Linkenkaer-Hansen, "Integrative EEG biomarkers predict
progression to Alzheimer's disease at the MCI stage," Frontiers in
Aging Neuroscience, vol. 5, 2013; G. Rodriguez, F. Nobili, A.
Arrigo et al., "Prognostic significance of quantitative
electroencephalography in Alzheimer patients: preliminary
observations," Electroencephalography and Clinical Neurophysiology,
vol. 99, no. 2, pp. 123-128, 1996; H. Soininen, J. Partanen, V.
Laulumaa, A. Paakkonen, E.-L. Helkala, and P. J. Riekkinen, "Serial
EEG in Alzheimer's disease: 3-year follow-up and clinical outcome,"
Electroencephalography and Clinical Neurophysiology, vol. 79, no.
5, pp. 342-348, 1991; P. Giannakopoulos, P. Missonnier, E. Kovari,
G. Gold, and A. Michon, "Electrophysiological markers of rapid
cognitive decline in mild cognitive impairment," Dementia in
Clinical Practice, vol. 24, pp. 39-46, 2009; P. M. Rossini, C. Del
Percio, P. Pasqualetti et al., "Conversion from mild cognitive
impairment to Alzheimer's disease is predicted by sources and
coherence of brain electroencephalography rhythms," Neuroscience,
vol. 143, no. 3, pp. 793-803, 2006; M. Buscema, E. Grossi, M.
Capriotti, C. Babiloni, and P. Rossini, "The I.F.A.S.T. model
allows the prediction of conversion to Alzheimer disease in
patients with mild cognitive impairment with high degree of
accuracy," Current Alzheimer Research, vol. 7, no. 2, pp. 173-187,
2010; L. S. Prichep, E. R. John, S. H. Ferris et al., "Prediction
of longitudinal cognitive decline in normal elderly with subjective
complaints using electrophysiological imaging," Neurobiology of
Aging, vol. 27, no. 3, pp. 471-481, 2006; K. M. Baerresen, K. J.
Miller, E. R. Hanson et al., "Neuro-psychological tests for
predicting cognitive decline in older adults," Neurodegenerative
Disease Management, vol. 5, no. 3, pp. 191-201, 2015; H. Brodaty,
M. H. Connors, D. Ames, and M. Woodward, "Progression from mild
cognitive impairment to dementia: a 3-year longitudinal study,"
Australian & New Zealand Journal of Psychiatry, vol. 48, no.
12, pp. 1137-1142, 2014; L. R. Clark, D. M. Schiehser, G. H.
Weissberger, D. P. Salmon, D. C. Delis, and M. W. Bondi, "Specific
measures of executive function predict cognitive decline in older
adults," Journal of the International Neuropsychological Society,
vol. 18, no. 1, pp. 118-127, 2012; P. Johnson, L. Vandewater, W.
Wilson et al., "Genetic algorithm with logistic regression for
prediction of progression to Alzheimer's disease," BMC
Bioinformatics, vol. 15, no. 16, 2014; T. Pereria, L. Lemos, S.
Cardoso et al., "Predicting progression of mild cognitive
impairment to dementia using neuropsychological data: a supervised
learning approach using time windows," BMC Med Inform Decision
Making, vol. 17, no. 1, 2017; H. Wilhalme, N. Goukasian, F. De Leon
et al., "A comparison of theoretical and statistically derived
indices for predicting cognitive decline," Alzheimer's &
Dementia: Diagnosis, Assessment & Disease Monitoring, vol. 6,
no. 1, pp. 171-181, 2017; C. Woolf, M. J. Slavin, B. Draper et al.,
"Can the clinical dementia rating scale identify mild cognitive
impairment and predict cognitive and functional decline?" Dementia
and Geriatric Cognitive Disorders, vol. 41, no. 5-6, pp. 292-302,
2016; J. Chung, E. Plitman, S. Nakajima et al., "Depressive
symptoms and small hippocampal volume accelerate the progression to
dementia from mild cognitive impairment," Journal of Alzheimer's
Disease: JAD, vol. 49, no. 3, pp. 743-754, 2015; S. Van der
Mussele, E. Fransen, H. Struyfs et al., "Depression in mild
cognitive impairment is associated with progression to Alzheimer's
disease: a longitudinal study," Journal of Alzheimer's Disease,
vol. 42, no. 4, pp. 1239-1250, 2014; A. G. Zippo and I.
Castiglioni, "Integration of (18) FDG-PET metabolic and functional
connectomes in the early diagnosis and prognosis of the Alzheimer's
disease," Current Alzheimer Research, vol. 13, no. 5, pp. 487-497,
2016; D. V. Moretti, G. B. Frisoni, C. Fracassi et al., "MCI
patients' EEGs show group differences between those who progress
and those who do not progress to AD," Neurobiology of Aging, vol.
32, no. 4, pp. 563-571, 2011; R. C. Petersen, G. E. Smith, S. C.
Waring, R. J. Ivnik, E. G. Tangalos, and E. Kokmen, "Mild cognitive
impairment: clinical characterization and outcome," Archives of
Neurology, vol. 56, no. 3, pp. 303-308, 1999; S. Gauthier, B.
Reisberg, M. Zaudig et al., "Mild cognitive impairment," The
Lancet, vol. 367, no. 9518, pp. 1262-1270, 2006. B. Reisberg, S.
Ferris, M. de Leon, and T. Crook, "The global deterioration scale
for assessment of primary degenerative dementia," The American
Journal of Psychiatry, vol. 139, no. 9, pp. 1136-1139, 1982; B.
Winblad, K. Palmer, M. Kivipelto et al., "Mild cognitive
impairment-beyond controversies, towards a consensus: report of the
international working group on mild cognitive impairment," Journal
of Internal Medicine, vol. 256, no. 3, pp. 240-246, 2004; A.
Hammers, R. Allom, M. J. Koepp et al., "Three-dimensional maximum
probability atlas
of the human brain, with particular reference to the temporal
lobe," Human Brain Mapping, vol. 19, no. 4, pp. 224-247, 2003; G.
Zhao and M. Pietikainen, "Dynamic texture recognition using local
binary patterns with an application to facial expressions," IEEE
Transactions on Pattern Analysis and Machine Intelligence, vol. 29,
no. 6, pp. 1-14, 2007; S. G. Mallat, "A theory for multiresolution
signal decom-position: the wavelet representation," IEEE
Transactions on Pattern Analysis and Machine Intelligence, vol. 11,
no. 7, pp. 674-693, 1989; M. J. Shensa, "The discrete wavelet
transforms: wedding a trous and mallat algorithms," IEEE
Transactions on Signal Processing, vol. 40, no. 10, pp. 2464-2482,
1992; M. N. Do and M. Vetterli, "Wavelet-based texture retrieval
using generalized Gaussian density and kullbackleibler distance,"
IEEE Transactions on Image Processing, vol. 11, no. 2, pp. 146-158,
2002; M. von Aster, A. Neubauer, and R. Horn, "Wechsler
intelligenztest fur Erwachsene WIE," in Deutschsprachige
Bearbeitung und Adaptation des WAIS-III von David Wechsler, Pearson
Assessment, Frankfurt, Germany, 2nd edition, 2006; C. Riekkinen and
H. F. Durwen, "VLMT: verbaler lern-und merkfa higkeitstest. ein
praktikables und differenziertes instrumentarium zur prufung der
verbalen gedachtnisleis-tungen," Schweiz Arch Neurol Psychiatr,
vol. 141, pp. 21-30, 1990; S. Weidlich and G. Lamberti,
Diagnosticum fur Cerebralsch Adigung: DCS; nach F. Hillers;
Handbuch, Huber Verlag, Bern, Vienna, Austria, 1980; A.
Aschenbrenner, O. Tucha, and K. Lange, RWT Regens-burger
Wortflussigkeits-Test. Handanweisung, Hogrefe Ver-lag, Gottingen,
Germany, 2000; P. Zimmermann and B. Fimm, A Test Battery for
Attentional Performance, pp. 110-151, Psychology Press, London, U
K, 2002; M. Hautzinger, F. Keller, and C. Kuhner, Beck Depressions
Inventar: Revision (BDI-II), Harcourt Test Services, Frankfurt,
Germany, 2006; Z. S. Nasreddine, N. A. Phillips, V. B cdirian et
al., "The montreal cognitive assessment, MoCA: a brief screening
tool for mild cognitive impairment," Journal of the American
Geriatrics Society, vol. 53, no. 4, pp. 695-699, 2005; A. Schlogl
and C. Brunner, "BioSig: a free and open-source software library
for BCI research," Computer, vol. 41, no. 10, pp. 44-50, 2008; S.
Marple, Digital Spectral Analysis with Applications, Prentice Hall,
Upper Saddle River, N.J., USA, 1987; V. K. Murthy, "Estimation of
the cross-spectrum," The Annals of Mathematical Statistics, vol.
34, no. 3, pp. 1012-1021, 1963; M. Kaminski, M. Ding, W. A.
Truccolo, and S. L. Bressler, "Evaluating causal relations in
neural systems: granger causality, directed transfer function and
statistical assessment of significance," Biological Cybernetics,
vol. 85, no. 2, pp. 145-157, 2001; M. Eichler, "On the evaluation
of information flow in multivariate systems by the directed
transfer function," Biological Cybernetics, vol. 94, no. 6, pp.
469-482, 2006; G. Nolte, O. Bai, L. Wheaton, Z. Mari, S. Vorbach,
and M. Hallett, "Identifying true brain interaction from EEG data
using the imaginary part of coherency," Clinical Neuro-physiology,
vol. 115, no. 10, pp. 2292-2307, 2004; W. Gersch and G. V. Goddard,
"Epileptic focus location: spectral analysis method," Science, vol.
169, no. 3946, pp. 701-702, 1970; L. A. Baccala and K. Sameshima,
"Partial directed coherence: a new concept in neural structure
determination," Biological Cybernetics, vol. 84, no. 6, pp.
463-474, 2001; L. Baccala, D. Takahashi, and K. Sameshima,
"Generalized partial directed coherence," in Proceedings of the
15th In-ternational Conference on Digital Signal Processing (DSP),
S. Sanei, J. Chambers, J. McWhirter et al., Eds., pp. 162-166 pp.
162-, Wales, UK, July 2007; M. J. Kaminski and K. J. Blinowska, "A
new method of the description of the information flow in the brain
structures," Biological Cybernetics, vol. 65, no. 3, pp. 203-210,
1991; A. Korzeniewska, M. Ma czak, M. Kami ski, K. J. Blinowska,
and S. Kasicki, "Determination of information flow direction among
brain structures by a modified directed transfer function (dDTF)
method," Journal of Neuroscience Methods, vol. 125, no. 1-2, pp.
195-207, 2003; S. L. Bressler, C. G. Richter, Y. Chen, and M. Ding,
"Cortical functional network organization from autoregressive
modeling of local field potential oscillations," Statistics in
Medicine, vol. 26, no. 21, pp. 3875-3885, 2007; J. Geweke,
"Measurement of linear dependence and feedback between multiple
time series," Journal of the American Statistical Association, vol.
77, no. 378, pp. 304-313, 1982; A. C. Bathke, S. Friedrich, M.
Pauly et al., "Testing mean differences among groups: multivariate
and repeated measures analysis with minimal assumptions,"
Multivariate Behavioral Research, vol. 53, no. 3, pp. 348-359,
2018; S. Friedrich, F. Konietschke, and M. Pauly, "MANOVA.RM: a
package for calculating test statistics and their resampling
versions for heteroscedastic semi-parametric multivariate data or
repeated measures designs," 2017; C.-Y. Wee, P.-T. Yap, and D.
Shen, "Prediction of Alzheimer's disease and mild cognitive
impairment using cortical morphological patterns," Human Brain
Mapping, vol. 34, no. 12, pp. 3411-3425, 2013; Y. Cui, P. S.
Sachdev, D. M. Lipnicki et al., "Predicting the development of mild
cognitive impairment: a new use of pattern recognition,"
Neuroimage, vol. 60, no. 2, pp. 894-901, 2012; X. Da, J. B. Toledo,
J. Zee et al., "Integration and relative value of biomarkers for
prediction of MCI to ad progression: spatial patterns of brain
atrophy, cognitive scores, apoe genotype and CSF biomarkers,"
NeuroImage: Clinical, vol. 4, pp. 164-173, 2014; C. Eckerstrom, E.
Olsson, M. Bjerke et al., "A combination of neuropsychological,
neuroimaging, and cerebrospinal fluid markers predicts conversion
from mild cognitive impairment to dementia," Journal of Alzheimer's
Disease, vol. 36, no. 3, pp. 421-431, 2013; S. C. Egli, D. I.
Hirni, K. I. Taylor et al., "Varying strength of cognitive markers
and biomarkers to predict conversion and cognitive decline in an
early-stage-enriched mild cognitive impairment sample," Journal of
Alzheimer's Disease, vol. 44, no. 2, pp. 625-633, 2015; S. J.
Buchert, J. Kurth, B. Krause, and M. J. Grothe, "Alzheimer's
disease neuroimaging initiative, 2015. The relative importance of
imaging markers for the prediction of Alzheimer's disease dementia
in mild cognitive impairment-beyond classical regression,"
NeuroImage: Clinical, vol. 8, pp. 583-593, 2015; K. van der Hiele,
E. L. E. M. Bollen, A. A. Vein et al., "EEG markers of future
cognitive performance in the elderly," Journal of Clinical
Neurophysiology, vol. 25, no. 2, pp. 83-89, 2008; D. Ferrazzoli, M.
Albanese, F. Sica et al., "Electroencephalography and dementia: a
literature review and future perspectives," CNS & Neurological
Disorders--Drug Targets, vol. 12, no. 4, pp. 512-519, 2013; N.
Hantke, K. A. Nielson, J. L. Woodard et al., "Comparison of
semantic and episodic memory BOLD fMRI activation in predicting
cognitive decline in older adults," Journal of the International
Neuropsychological Society, vol. 19, no. 1, pp. 11-21, 2013; N. A.
Kochan, M. Breakspear, M. Valenzuela et al., "Cortical responses to
a graded working memory challenge predict functional decline in
mild cognitive impairment," Biological Psychiatry, vol. 70, no. 2,
pp. 123-130, 2011; S. L. Miller, E. Fenstermacher, J. Bates, D.
Blacker, R. A. Sperling, and B. C. Dickerson, "Hippocampal
activation in adults with mild cognitive impairment predicts
subsequent cognitive decline," Journal of Neurology, Neurosurgery
& Psychiatry, vol. 79, no. 6, pp. 630-635, 2007; S. Haller, D.
Nguyen, C. Rodriguez et al., "Individual prediction of cognitive
decline in mild cognitive impairment using support vector
machine-based analysis of diffusion tensor imaging data," Journal
of Alzheimer's Disease, vol. 22, no. 1, pp. 315-327, 2010; M. A.
Lancaster, M. Seidenberg, J. C. Smith et al., "Diffusion tensor
imaging predictors of episodic memory decline in healthy elders at
genetic risk for Alzheimer's disease," Journal of the International
Neuropsychological Society, vol. 22, no. 10, pp. 1005-1015, 2016;
S.-H. Jin and C. K. Chung, "Functional substrate for memory
function differences between patients with left and right mesial
temporal lobe epilepsy associated with hippocampal sclerosis,"
Epilepsy & Behavior, vol. 51, pp. 251-258, 2015. G. E. Doucet,
X. He, M. Sperling, A. Sharan, and J. I. Tracy, "Gray matter
abnormalities in temporal lobe epilepsy: relationships with
resting-state functional connectivity and episodic memory
performance," PLoS One, vol. 11, no. 5, Article ID e0154660, 2016;
V. Dinkelacker, R. Valabregue, L. Thivard et al.,
"Hippocampal-thalamic wiring in medial temporal lobe epilepsy:
enhanced connectivity per hippocampal voxel," Epilepsia, vol. 56,
no. 8, pp. 1217-1226, 2015; A. J. Watrous, N. Tandon, C. R. Conner,
T. Pieters, and A. D. Ekstrom, "Frequency-specific network
connectivity increases underlie accurate spatiotemporal memory
retrieval," Nature Neuroscience, vol. 16, no. 3, pp. 349-356, 2013;
D. W. Zaidel, M. M. Esiri, and J. M. Oxbury, "Regional
differentiation of cell densities in the left and right hippocampi
of epileptic patients," Journal of Neurology, vol. 240, no. 5, pp.
322-325, 1993; E. Moradi, A. Pepe, C. Gaser, H. Huttunen, and J.
Tohka, "Machine learning framework for early MRI-based Alzheimer's
conversion prediction in MCI subjects," Neuro-image, vol. 104, pp.
398-412, 2015; O. Hardt, K. Nader, and L. Nadel, "Decay happens:
the role of active forgetting in memory," Trends in Cognitive
Sciences, vol. 17, no. 3, pp. 111-120, 2013; S. J. Wakefield, D. J.
Blackburn, K. Harkness, A. Khan, M. Reuber, and A. Venneri,
"Distinctive neuropsychological profiles differentiate patients
with functional memory dis-order from patients with amnestic-mild
cognitive impairment," Acta Neuropsychiatrica, vol. 30, no. 2, pp.
90-96, 2018; M. Gschwandtner, Y. Holler, M. Liedlgruber, E. Trinka,
and A. Uhl, "Assessing out-of-the-box software for automated
hippocampus segmentation," in Bildverarbeitung fur die Medizin
2016: Algorithmen--Systeme--Anwendungen, T. Tolxdorff, T. M.
Deserno, H. Handels et al., Eds., Springer Berlin Heidelberg,
Berlin, Germany, pp. 212-217, 2016; M. A. Araque Cabellero, S.
Kloppel, M. Dichgans, and M. Ewers, "Spatial patterns of
longitudinal gray matter change as predictors of concurrent
cognitive decline in amyloid positive healthy subjects," Journal of
Alzheimer's Disease, vol. 55, no. 1, pp. 343-358, 2017; R. A.
Sarkis, B. C. Dickerson, A. J. Cole, and Z. N. Chemali, "Clinical
and neurophysiologic characteristics of unprovoked seizures in
patients diagnosed with dementia," The Journal of Neuropsychiatry
and Clinical Neurosciences, vol. 28, no. 1, pp. 56-61, 2016; Olaf
Ronneberger, Philipp Fischer, and Thomas Brox, "U-Net:
Convolutional Networks for Biomedical Image Segmentation", 18 May
2015; Francois Lazeyras, PhD, Olaf Blanke, Steven Perrig, Ivan
Zimine, Xavier Golay, Jacqueline Delavelle, Christoph M. Michel,
Nicolas de Tribolet, Jean-Guy Villemure, and Margitta Seeck;
"EEG-Triggered Functional MRI in Patients with Pharmacoresistant
Epilepsy", 2000; Diedre Carmo, Leticia Rittner, Roberto Lotufo,
Bruna Silva, Clarissa Yasuda, "Extended 2D Consensus Hippocampus
Segmentation", 2019; Giuseppe Palma, Enrico Tedeschi, Pasquale
Borrelli, Sirio Cocozza, Carmela Russo, Saifeng Liu, Yongquan Ye,
Marco Comerci, Bruno Alfano, Marco Salvatore, E. Mark Haacke,
Marcello Mancini, "A Novel Multiparametric Approach to 3D
Quantitative MRI of the Brain;" Susan M. Resnick, Dzung L. Pham,
Michael A. Kraut, Alan B. Zonderman, and Christos Davatzikos,
"Longitudinal Magnetic Resonance Imaging Studies of Older Adults: A
Shrinking Brain"; Apr. 15, 2003; Azar Zandifara, Vladimir Fonova,
Pierrick Coupec, Jens Pruessnerd, D. Louis Collinsa, "A comparison
of accurate automatic hippocampal segmentation methods for the
Alzheimer's Disease Neuroimaging Initiative;" V. Calhoun, T. Adal,
and J. Liu, "A feature-based approach to combine functional MRI,
structural MRI and EEG brain imaging data", 2006; Erik K. St.
Louis, Lauren C. Frey, "An Introductory Text and Atlas of Normal
and Abnormal Findings in Adults, Children, and Infants", N. White,
S. Magda, C. Airriess2, J. Albright, "The New Personalized
Segmentation Approach"; Kerry W. Kilborn, Zoe Tieges, Jessica
Price, Susil Stephen, Bernard A. Conway, Delphine Duclap, Alan H.
Hughes, Gillian McLean, "Source Localization of Event-Related
Potential effects differentiates between vascular dementia and
Alzheimer's disease;" Saman Sargolzaei, Arman Sargolzaei, Mercedes
Cabrerizo, Gang Chen, Mohammed Goryawala, Alberto Pinzon-Ardila,
Sergio M. Gonzalez-Ariaa, Malek Adjouadi, "Estimating Intracranial
Volume in Brain Research: an Evaluation of Methods;" B. BRENT
SIMMONS, and BRETT HARTMANN, "Evaluation of Suspected Dementia;"
DAVID R. FISH AND SUSAN S. SPENCER, "Clinical Correlations: MRI AND
EEG;" Emma R. Mulder, Remko A. de Jong, Dirk L. Knol, Ronald A. van
Schijndel, Keith S. Cover, Pieter J. Visser, Frederik Barkhof, Hugo
Vrenken, "Hippocampal volume change measurement: Quantitative
assessment of the reproducibility of expert manual outlining and
the automated methods for the Alzheimer's Disease Neuroimaging
Initiative", 2014; Dorothee Schoemaker, Claudia Buss, Kevin Head,
Curt A. Sandman, Elysia P. Davis, M. Mallar Chakravarty, Serge
Gauthier a, Jens C. Pruessner, "Hippocampus and amygdala volumes
from magnetic resonance images in children: Assessing accuracy of
FreeSurfer and FSL against manual segmentation; Francois De Guio,
Marco Duering, Franz Fazekas, Frank-Erik De Leeuw, Steve Greenberg,
Leonardo Pantoni, Agne's Aghetti, Eric E Smith, Joanna Wardlaw, and
Eric Jouvent, "Brain atrophy in cerebral small vessel diseases:
Extent, consequences, technical limitations and perspectives: The
HARNESS initiative;" Charles W. Kanaly, Dale Ding, Ankit I. Mehta,
Anthony F. Waller, Ian Crocker, Annick Desjardins, David A.
Reardon, Allan H. Friedman, Darell D. Bigner, and John H. Sampson,
"A Novel Method for Volumetric MRI Response Assessment of Enhancing
Brain Tumors;" Simon S. Keller, Jan-Christoph Schoene-Bake, Jan S.
Gerdes, Bernd Weber, Michael Deppe, "Concomitant Fractional
Anisotropy and Volumetric Abnormalities in Temporal Lobe Epilepsy:
Cross-Sectional Evidence for Progressive Neurologic Injury;" Tomas
Kalincik, Manuela Vaneckova, Michaela Tyblova, Jan Krasensky,
Zdenek Seidl, Eva Havrdova, Dana Horakova, "Volumetric MRI Markers
and Predictors of Disease Activity in Early Multiple Sclerosis: A
Longitudinal Cohort Study;" Yu Zhang, Norbert Schuff, Monica
Camacho, Linda L. Chao, Thomas P. Fletcher, Kristine Yaffe, Susan
C. Woolley, Catherine Madison, Howard J. Rosen, Bruce L. Miller,
Michael W. Weiner; "MRI Markers for Mild Cognitive Impairment:
Comparisons between White Matter Integrity and Gray Matter Volume
Measurements;" Ricardo Saute, Kevin Dabbs, Jana E. Jones, Daren C.
Jackson, Michael Seidenberg, Bruce P. Hermann, "Brain Morphology in
Children with Epilepsy and ADHD;" Margaret R. Lentz, Kristin L.
Peterson, Wael G. Ibrahim, Dianne E. Lee, Joelle Sarlls, Martin J.
Lizak, Dragan Maric, William C. Reid, Dima A. Hammoud, "Diffusion
Tensor and Volumetric Magnetic Resonance Measures as Biomarkers of
Brain Damage in a Small Animal Model of HIV;" Sanaz Gabery, Nellie
Georgiou-Karistianis, Sofia Hult Lundh, Rachel Y. Cheong, Andrew
Churchyard, Phyllis Chua, Julie C. Stout, Gary F. Egan, Deniz
Kirik, .ANG.sa Petersen "Volumetric Analysis of the Hypothalamus in
Huntington Disease Using 3T MRI: The IMAGE-HD Study;" Jose Carlos
Delgado-Gonzalez, Francisco Mansilla-Legorburo, Jose Florensa-Vila,
Ana Maria Insausti, Antonio Vinuela, Teresa Tunon-Alvarez, Marcos
Cruz, Alicia Mohedano-Moriano, Ricardo Insausti, Emilio
Artacho-Perula, "Quantitative Measurements in the Human Hippocampus
and Related Areas: Correspondence between Ex-Vivo MRI and
Histological Preparations;" Kashif Rajpoot, Atif Riaz, Waqas
Majeed, Nasir Rajpoot, "Functional Connectivity Alterations in
Epilepsy from Resting-State Functional MRI;" Rene-Maxime Gracien,
Alina Jurcoane, Marlies Wagner, Sarah C. Reitz, Christoph Mayer,
Steffen Volz, Stephanie-Michelle Hof, Vinzenz Fleischer, Amgad
Droby, Helmuth Steinmetz, Frauke Zipp, Elke Hattingen, Ralf
Deichmann, Johannes C. Klein, "The Relationship between Gray Matter
Quantitative MRI and Disability in Secondary Progressive Multiple
Sclerosis;" Meriem El Azami, Alexander Hammers, Julien Jung,
Nicolas Costes, Romain Bouet, Carole Lartizien, "Detection of
Lesions Underlying Intractable Epilepsy on T1-Weighted MRI as an
Outlier Detection Problem;" Oeslle Lucena, Roberto Souza, Leticia
Rittner, Richard Frayne, Roberto Lotufo, "SILVER STANDARD MASKS FOR
DATA AUGMENTATION APPLIED TO DEEP-LEARNING-BASED SKULL-STRIPPING",
2018; William J. McGeown, Marco Cecchi, K C Fadem,
"NEUROPSYCHOLOGICAL AND NEUROANATOMICAL CORRELATES OF EVENT-RELATED
POTENTIALS IN PATIENTS WITH ALZHEIMER'S DISEASE;" Nikdokht Farid,
Holly M. Girard, Nobuko Kemmotsu, Michael E. Smith, Sebastian W.
Magda, Wei Y. Lim, Roland R. Lee, Carrie R. McDonald, "Temporal
Lobe Epilepsy: Quantitative MR Volumetry in Detection of
Hippocampal Atrophy;" Bahman Nasseroleslami, Stefan Dukic, Michael
Broderick, Kieran Mohr, Christina Schuster, Brighid Gavin, Russell
McLaughlin, Mark Heverin, Alice Vajda, Parameswaran M. Iyer, Niall
Pender, Peter Bede, Edmund C. Lalor, and Orla
Hardiman, "Characteristic Increases in EEG Connectivity Correlate
with Changes of Structural MRI in Amyotrophic Lateral Sclerosis;"
D. Heister, J. B. Brewer, S. Magda, et al., "Predicting MCI outcome
with clinically available MRI and CSF Biomarkers", 2011; Trygve B.
Leergaard, Nathan S. White, Alex de Crespigny, Ingeborg Bolstad,
Helen D'Arceuil, Jan G. Bjaalie, Anders M. Dale, "Quantitative
Histological Validation of Diffusion MRI Fiber Orientation
Distributions in the Rat Brain;" Riccardo Metere, Tobias Kober,
Harald E. Moller, Andreas Schafer, "Simultaneous Quantitative MRI
Mapping of T1, T2 and Magnetic Susceptibility with Multi-Echo
MP2RAGE;" Xin Xu, Jinshan Tang, Xiaolong Zhang, Xiaoming Liu, Hong
Zhang, and Yimin Qiu, "Exploring Techniques for Vision Based Human
Activity Recognition: Methods, Systems, and Evaluation", 2013; Dr
Subhash Kaul, "Stroke in India;" Benjamin Thyreau, Kazunori Sato,
Hiroshi Fukuda, Yasuyuki Taki, "Segmentation of the Hippocampus by
Transferring Algorithmic Knowledge for large cohort processing;"
Woo Suk Tae, Sam Soo Kim, Kang Uk Lee, Eui-Cheol Nam, and Keun Woo
Kim, "Validation of Hippocampal volumes measured using a manual
method and two automated methods in chronic major depressive
disorder;" Junko Matsuzawa, Mie Matsui, Tohru Konishi, Kyo Noguchi,
Ruben C. Gur, Warren Bilker and Toshio Miyawaki; "Age-related
Volumetric Changes of Brain Gray and White Matter in Healthy
Infants and Children;" Michael Wagner, and Manfred Fuchs,
"Integration of Functional MRI, Structural MRI, EEG, and MEG," Sven
Haller, Eniko Kovari, Francois R Herrmann, Victor Cuvinciuc,
Ann-Marie Tomm, Gilbert B Zulian, Karl-Olof Lovblad, Panteleimon
Giannakopoulos, and Constantin Bouras, "Do brain T2/FLAIR white
matter hyperintensities correspond to myelin loss in normal aging?
A radiologic-neuropathologic correlation study;" U.S. Pat. No.
7,283,652B2 entitled "Method and system for measuring disease
relevant tissue changes;" US20100266173 entitled "Computer-aided
detection (cad) of a disease;" U.S. Pat. No. 9,588,204 entitled
"Magnetic resonance spectroscopic imaging volume of interest
positioning;" US20080249396 entitled "Method and Apparatus for
Determining Indications Helping the Diagnosis of Orthopedical
Diseases;" US20190109830 entitled "Systems and Methods for Ensuring
Data Security in the Treatment of Diseases and Disorders Using
Digital Therapeutics;" US20190139223 entitled "System and method
for extracting a region of interest from volume data;" U.S. Pat.
No. 6,901,280 entitled "Evaluating disease progression using
magnetic resonance imaging;" CN101593345 entitled
"Three-dimensional medical image display method based on the GPU
acceleration;" US20200315455 entitled "Medical image processing
system and method for personalized brain disease diagnosis and
status determination;" U.S. Pat. No. 8,634,614B2 entitled "System
and method for volumetric analysis of medical images;"
US20130267827 entitled "Method and magnetic resonance system for
functional MR imaging of a predetermined volume segment of the
brain of a living examination subject," US20100080432 entitled
"Tools for aiding in the diagnosis of neurodegenerative diseases";
and EP3714467 entitled "Content based image retrieval for lesion
analysis;" US20180220984A1 entitled "Medical Imaging Methods and
Apparatus for Diagnosis and Monitoring of Diseases and Uses
Therefor;" U.S. Ser. No. 10/878,219B2 entitled "Method and system
for artificial intelligence based medical image segmentation;"
JP5366356B2 entitled "Medical image processing apparatus and
medical image processing method;" US20040078238A1 entitled
"Anonymizing tool for medical data;" U.S. Ser. No. 10/198,832B2
entitled "Generalizable medical image analysis using segmentation
and classification neural networks;" US20180103917A1 entitled
"Head-mounted display EEG device."
* * * * *